Bimanual coordination: A missing piece of arm rehabilitation after stroke

Abstract

Inability to use the arm in daily actions significantly lowers quality of life after stroke. Most contemporary post-stroke arm rehabilitation strategies that aspire to re-engage the weaker arm in functional activities have been greatly limited in their effectiveness. Most actions of daily life engage the two arms in a highly coordinated manner. In contrast, most rehabilitation approaches predominantly focus on restitution of the impairments and unilateral practice of the weaker hand alone. We present a perspective that this misalignment between real world requirements and intervention strategies may limit the transfer of unimanual capability to spontaneous arm use and functional recovery. We propose that if improving spontaneous engagement and use of the weaker arm in real life is the goal, arm rehabilitation research and treatment need to address the coordinated interaction between arms in targeted theory-guided interventions. Current narrow focus on unimanual deficits alone, difficulty in quantifying bimanual coordination in real-world actions and limited theory-guided focus on control and remediation of different coordination modes are some of the biggest obstacles to successful implementation of effective interventions to improve bimanual coordination in the real world. We present a theory-guided taxonomy of bimanual actions that will facilitate quantification of coordination for different real-world tasks and provide treatment targets for addressing coordination deficits. We then present evidence in the literature that points to bimanual coordination deficits in stroke survivors and demonstrate how current rehabilitation approaches are limited in their impact on bimanual coordination. Importantly, we suggest theory-based areas of future investigation that may assist quantification, identification of neural mechanisms and scientifically-based training/remediation approaches for bimanual coordination deficits post-stroke. Advancing the science and practice of arm rehabilitation to incorporate bimanual coordination will lead to a more complete functional recovery of the weaker arm, thus improving the effectiveness of rehabilitation interventions and augmenting quality of life after stroke.

Keywords

Bimanual coordination training arm rehabilitation stroke

1 Introduction

Each year, stroke affects nearly 800,000 people in the United States, with many survivors experiencing persistent functional deficits and poor quality of life (Winstein et al., 2016). A vast majority of people with stroke experience weakness in the upper extremity (UE) that limits their ability to use their arm and hand in daily life (Faria-Fortini, Michaelsen, Cassiano, & Teixeira-Salmela, 2011; Wade, Langton-Hewer, Wood, Skilbeck, & Ismail, 1983). In particular, arm deficits are significant because 50% of the reduction in quality of life after stroke is due to the inability to use their arm effectively in actions of daily living (Kwakkel, Kollen, & Lindeman, 2004; Winstein et al., 2016). Indeed, a major rehabilitation goal of patients with stroke is to be able to recover arm weakness and re-engage the affected arm in functional activities. However, clinical efforts in improving arm function after stroke have had limited success, with only 5% of stroke survivors regaining full arm function despite extensive therapy (Sathian et al., 2011). These statistics attest to the need for innovative and more comprehensive strategies to improve arm function after stroke.

In neurologically-intact individuals, most functional activities are accomplished by using both hands in a highly coordinated and efficient manner (Bailey, Klaesner, & Lang, 2015a; Kilbreath & Heard, 2005). This bimanual engagement is different for different tasks; yet is performed in a seamless, spontaneous manner. After unilateral stroke, performance of most of these functional activities deteriorates and bilateral arm use is reduced compared to neurologically-intact individuals. Unilateral stroke often results in motor and sensory impairments that are more obvious in the contralesional weaker (paretic) arm (Beer, Dewald, & Rymer, 2000; Faria-Fortini, Michaelsen, Cassiano, & Teixeira-Salmela; Sathian et al., 2011; Takahashi & Reinkensmeyer, 2003; Wagner et al., 2006; Winstein et al., 2016). Therefore, it is not surprising that majority of UE rehabilitation strategies are predominantly focused on improving unimanual performance of the weaker arm (Conroy et al., 2011; Lo et al., 2010; Mazzoleni et al., 2013; Wolf, 2007; Wolf et al., 2006). The implicit assumption underlying predominantly unimanual approaches is that improving the performance of the weaker arm will automatically translate to improvements in their ability for bimanual control and coordination. However, there is mounting evidence that unimanual impairments alone may not be highly predictive of bimanual coordination deficits (Kantak, Zahedi, & McGrath, 2016; Lowrey & Jackson, 2014; Sainburg, Good, & Przybyla, 2013). In contrast, bimanual coordination after stroke is likely to be dependent upon a complex interaction between task requirements, lesion location, residual brain connectivity (e.g., interhemispheric connectivity), and sensorimotor impairments and performance of both paretic and nonparetic arms (Gerloff & Andres, 2002; Gooijers & Swinnen, 2014; Sainburg et al., 2013; Wolf et al., 2014).

In this perspective, our thesis is that the narrow focus of research and interventions on the paretic arm alone, with little regard to the complex, task-dependent interaction between hands may be an important limitation of arm rehabilitation after stroke. We propose that targeted training of coordinated interaction between the two arms is necessary for achieving more complete functional recovery and integration of the weaker arm in activities of daily living. In the first part, we highlight the role of bimanual movements in activities of daily living and review current evidence to identify deficits in bimanual coordination after unilateral stroke. Specifically, we suggest that theory-guided classification for bimanual actions may provide a framework to evaluate bimanual coordination deficits during daily actions, target treatment strategies and assess subsequent outcomes for bimanual actions. In the next section, we critically appraise the current rehabilitation strategies for their effects on bimanual coordination and arm use in daily life. Finally, we identify important gaps in the current rehabilitation research and propose areas of systematic investigation in order to advance the assessment and training of coordinated bimanual actions for a more comprehensive arm rehabilitation after stroke.

2 Bimanual coordination-essential aspect of human behavior

Converging evidence from early observational (Kilbreath & Heard, 2005) and recent studies with accelerometers (Bailey, Klaesner, & Lang, 2014; Bailey, Klaesner, et al., 2015a; Rand & Eng, 2010) in neurologically-intact individuals suggests almost equivalent use of the dominant and nondominant arm during a typical day. While performing bimanual actions of daily living, the two arms are engaged in a concerted spatial and temporal interrelationship that characterizes bimanual coordination (Howard, Ingram, Körding, & Wolpert, 2009). This coordination between the two upper extremities makes actions skillful because each arm contributes to an action component while interacting with the other in a fine spatial and temporal manner to ensure that the action is accurate and efficient (Franz, 2003). Collectively, these studies indicate that bimanual coordination is an important aspect of skilled arm use necessary for optimal human function.

Coordinating two limbs, each having multiple degrees of freedom, to accomplish an array of functional tasks under varied environmental conditions poses a clear challenge for the human sensori-motor system. Despite this challenge, task-oriented actions are highly coordinated such that common elements are tightly controlled in a healthy neurologically-intact people. Multiple theories of motor control have suggested that there are inherent constraints in the nervous system that limit the number of action choices (i.e. degrees of freedom) that yield coordinated movements. For example, dynamical systems theory proposes that coordination between the two hands (mostly during rhythmic movements) emerges from stability of interactions within the system (Kelso, 1983; Scholz, 1990). Experimentally, the so-called “self-organization” of motor behavior that occurs through the non-linear dynamics of the interacting elements of the system (Jirsa, Fuchs, & Kelso, 1998), may be addressed through measurement of movement parameters. These include order parameters (e.g., symmetry of movement) and controls parameters (e.g., speed of movements). In contrast muscle synergy theories suggest that a few set of muscle/movement combinations limit the number of solutions to the degree of freedom problem and result in multiple coordinated synergies (d’Avella & Lacquaniti, 2013; Jarrassé, Ribeiro, Sahbani, Bachta, & Roby-Brami, 2014; Vinjamuri, Sun, Crammond, Sclabassi, & Mao, 2008). In contrast to these theories, the internal model theory posits that a central representation (i.e., an internal model) for bimanual actions that is modified based on the motor, cognitive and perceptual constraints imposed by the task and environment may help determine the nature of coordination between arms (Yokoi, Hirashima, & Nozaki, 2011). Finally, recent theoretical frameworks (e.g., Optimal feedback control theory (Diedrichsen, Shadmehr, & Ivry, 2010)) propose that in addition to central processes, task-dependent sensory feedback optimizes motor strategy by sharing different action components between the two hands such that the emergent coordination for a task goal minimizes energy costs and makes the movements more efficient. While all these theories have important implications on bimanual coordination in patients with stroke, the scope of the current perspective does not permit a detailed elaboration of these aforementioned theories, which are not necessarily exclusive of each other, and all of which consider localization of their elements in the central nervous system.

Despite their differences, all theories concur that the nature of coordination between arms is dependent upon the requirements of the task. While there are multitude of ways in which different tasks engage the two arms, the spatiotemporal characteristics of this engagement can be classified based on the symmetry of arm movements (MacKenzie & RG, 1985) and the nature of how the task goals are conceptualized (Diedrichsen & Gush, 2009; Kantak, McGrath, & Zahedi, 2016; Mechsner, Knoblich, & Mechsner, 2004). Figure 1 provides a classification of bimanual actions that is based on two principal theoretical determinants of bimanual coordination: symmetry of arm movements and conceptualization of task goals. Building upon the “coordination modes” in motor control research (Kelso, 1983; MacKenzie & RG, 1985), the present classification provides a framework upon which daily bimanual actions can be categorized into theoretically-based classes of bimanual coordination. Symmetric bimanual actions simultaneously engage homologous muscles while asymmetric movements engage non-homologous muscles either simultaneously or in different temporal relationships. Motor control studies in neurologically-intact individuals and patients with neurological injuries have used different laboratory and real-world tasks to characterize bimanual coordination. Relatively large proportion of these studies have demonstrated that symmetric bimanual movements are relatively simpler and “more stable” than asymmetric movements (Kelso, 1983; Lewis & Byblow, 2004; Swinnen, 2002; Swinnen & Wenderoth, 2004). While majority of the studies have employed rhythmic bimanual movements, relatively fewer have investigated discrete or functional bimanual actions. Asymmetric bimanual movements have been shown to engage greater processing during response preparation and execution (Blinch, Cameron, Franks, Carpenter, & Chua, 2015; Blinch, Cameron, Franks, & Chua, 2011). Not surprisingly, the behavioral differences between symmetric and asymmetric movements are supported by distinct neural networks. Compared to symmetric movements, there is greater activation of motor association areas such as supplementary motor area and right dorsal premotor area during asymmetric bimanual movements (Aramaki, Osu, & Sadato, 2010; Maki, Wong, Sugiura, Ozaki, & Sadato, 2008). Also, symmetric and asymmetric actions are associated with distinct changes in interhemispheric communication. In neurologically-intact individuals, symmetric bimanual movements were associated with greater transcallosal inhibition (TCI) compared to asymmetric bimanual movements (Long, Tazoe, Soteropoulos, & Perez, 2016; Perez, Butler, & Taylor, 2014; Tazoe, Sasada, Sakamoto, & Komiyama, 2013). Thus symmetry of arm movements is a critical distinctive factor that influences behavioral and neurophysiologic control of bimanual actions, and hence is included in the classification.

Fig.1

Classification of bimanual actions based on symmetry of arm movements and conceptualization of task goals.

Besides the motor factors that influence bimanual coordination, how one perceives the task goals is crucial in determining the coordination between the two arms (Diedrichsen, Grafton, Albert, Hazeltine, & Ivry, 2006a; Diedrichsen & Gush, 2009; Mechsner, Kerzel, Knoblich, & Prinz, 2001; Shea, Buchanan, & Kennedy, 2016). During some bimanual actions, each arm moves to accomplish independent goals (e.g. one hand lifting a cup and the other hand lifting a glass simultaneously) while in other instances both hands work together to accomplish a common task goal (e.g. lifting a tray). Coordination during common goal actions may further differ depending on the necessity of cooperative interaction between the two hands. In some tasks classified here as parallel tasks, such as opening a drawer to grab a pen, the two hands move almost simultaneously – one to open the drawer and the other to reach to grab the pen. This simultaneity of arm movements, although desirable for efficiency, is not necessary for task accomplishment. In other words, one can open the drawer first and subsequently reach to grab the pen after the drawer is opened. Thus, in parallel tasks, the spatiotemporal interrelationship is desirable for efficiency, but not necessary for task success. In contrast, cooperative tasks are those common-goal tasks where cooperative interaction between the two arms is necessary for task success (e.g. picking up a box). To pick a box, both hands must synchronously apply forces to grip and lift the box, while compensating for, and co-varying the forces to ensure smooth pick up and move. If one of the arms is unable to cooperate with the other arm, the task may not be accomplished successfully. Thus during cooperative tasks, the spatiotemporal interaction between hands/arms is necessary for task success. This is true for both symmetric actions such as picking a box, as well as asymmetric cooperative actions such as cutting a piece of steak with a knife in one hand while the other hand stabilizes the steak with a fork.

So far, bimanual coordination research in basic motor control and clinical assessment of functional bimanual activities has progressed without a common theory-guided taxonomy (MacKenzie & RG, 1985; Shirota et al., 2016). While it is agreed upon that the performance of functional tasks depends on neuromotor control, lack of a common taxonomy bridging motor control constructs and clinical assessments poses an important limitation in the scientific progress of behavioral, neurophysiological and clinical investigation of bimanual deficits. The proposed classification provides a theory-based categorization upon which different functional tasks can be mapped. The classification will allow testing of functional tasks and likely provide insight into the “mode” of coordination that is impaired in specific patients. While the proposed classification requires rigorous testing across laboratory-based and real-world tasks in individuals with and without bimanual coordination deficits, we believe this initial attempt to classify bimanual actions into theoretically-based categories is important to move the field of bimanual coordination rehabilitation forward.

Comprehensive understanding of upper extremity dysfunction following stroke requires identifying post-stroke changes in bimanual coordination, factors that modulate (impair or improve) coordination, patient characteristics that contribute to coordination impairments as well as behavioral and functional consequences of altered coordination. In the following section, we outline studies that have investigated the changes in bimanual coordination after unilateral stroke.

3 Bimanual coordination after unilateral stroke

After unilateral stroke, it is not surprising that the activity of the paretic arm is greatly reduced during daily activities (Bailey, Birkenmeier, & Lang, 2015; Bailey, Klaesner, & Lang, 2015). Importantly, accelerometry studies have also reported reduced intensity of bilateral upper limb activity, with greater contribution of the nonparetic arms during bimanual actions in patients with stroke compared to age-matched controls (Haaland et al., 2012; Michielsen, Selles, Stam, Ribbers, & Bussmann, 2012). The reduced use of the weaker hand in unimanual and bimanual actions has been traditionally thought to be a result of impaired unimanual performance (Beer et al., 2000; Roh, Rymer, & Beer, 2015; Roh, Rymer, Perreault, Yoo, & Beer, 2013; Wagner et al., 2006). Recent data has challenged the notion. Waddell and colleagues reported that despite significant improvements in paretic arm capability (tested using the Action Research Arm Test, ARAT) following intensive upper limb intervention, paretic arm activity outside the clinic did not improve significantly (Waddell et al., 2016). This counter-intuitive finding suggests that improving unimanual performance may not be sufficient to promote a more functional upper extremity function after stroke. An important, yet less studied factor that likely determines spontaneous arm use in daily life after stroke may be how efficiently, simultaneously and skillfully the two arms coordinate with each other.

Bimanual coordination in patients with stroke has been broadly studied at two levels of the translational pipeline. A more basic behavioral science level employs well-controlled laboratory-based kinematic and kinetic tasks while a more applied approach uses naturalistic tasks to study the coordination between arms. Basic behavioral studies offer important insights into the fundamental behavioral mechanisms underlying bimanual coordination, while the more applied naturalistic task studies offer external validity and identify coordination deficits more directly relevant to rehabilitation. Almost all of the studies discussed below have investigated bimanual coordination in patients with chronic stroke, with one exception. Metrot and colleagues investigated changes in bimanual coordination during a bilateral reach-to-grasp task in a cohort of 12 moderately hemiparetic patients within a month of stroke. Patients were followed up for 6 weeks and at 3 months after inclusion. Bimanual coordination improved over the first 6 weeks after stroke, as evidenced by improved kinematics and reduced variability of between-hand synchronization. It is however open to question if the improvements in bimanual coordination perfectly mirror those of the paretic arm and are generalizable across different motor tasks.

3.1 Kinematic studies of bimanual coordination in post-stroke patients

Scientists have long recognized the inherent linkages that bias the symmetry of bilateral movements. This recognition, driven through the tenets of the dynamical systems theory, led to a substantial body of literature that determined the effects of symmetrical independent-goal bimanual movements on the performance of the paretic arm in patients with stroke. As early as 1951, Cohn reported that movement of the weaker side reduced the frequency of the unimpaired forearm during bilateral continuous pronation-supination movements (Cohn et al., 1951). These findings, in addition to subsequent studies using an elbow flexion-extension task (Cunningham, Stoykov, & Walter, 2002), continuous circle drawing task (Lewis & Byblow, 2004), bimanual reaching task (McCombe Waller, Harris-Love, Liu, & Whitall, 2006; Rose & Winstein, 2005) and bilateral reach-to-grasp tasks (Mudie & Matyas, 2000) indicate that despite unimanual weakness, patients demonstrate a capability to retain spatial and temporal coupling between the two UEs. Particularly, during symmetric independent-goal and common-goal parallel tasks, the paretic arm speeds up while the nonparetic arm slows down to retain temporal coupling between arms. This “facilitatory or coupling” effect of the bimanual symmetric condition on paretic arm performance has formed the bases for many therapeutic approaches to improve paretic arm movements (Arya & Pandian, 2014; Daly, Harris-Love, Waller, & Whitall, 2005; Sleimen-Malkoun, Temprado, Thefenne, & Berton, 2011; Stinear, Barber, Coxon, Fleming, & Byblow, 2008; Whitall, McCombe Waller, Silver, & Macko, 2000a). In contrast to symmetric movements to independent goals, there is evidence that bimanual coordination is impaired when patients perform asymmetric actions (Rose & Winstein, 2013).

Much work discussed above has focused on tasks that required each hand to accomplish an independent goal with a focus on how performance of one hand influences that of the other (Dejong & Lang, 2012; McCombe Waller et al., 2006). These studies have provided important insights into the inherent behavioral and neurophysiological linkages between the two arms and the effects of stroke on those linkages. Further, most studies employed rhythmic bimanual movements, rather than activities of daily living. Most real-world actions require the two hands to override the inherent linkages in order to work in a cooperative manner to accomplish a common goal (Diedrichsen, Grafton, Albert, Hazeltine, & Ivry, 2006b; Julie Duque et al., 2010; Kantak, McGrath, & Zahedi, 2016). During such cooperative movements, the two arms interact with each other, compensating for, and covarying with, each other to successfully accomplish the task. Imaging studies have demonstrated that common goal actions engage a distinct neural substrate from independent goal actions in neurologically-intact individuals (Duque et al., 2010). In the referenced study, during common-goal condition, there is a greater activation in the right superior temporal gyrus and supplementary motor area compared to the independent-goal condition. Further, cooperative coordination is impaired when right superior temporal gyrus is transiently perturbed using transcranial magnetic stimulation, suggesting a causative link between right STG function and cooperative bimanual coordination. Little is known how common goal cooperative bimanual coordination is affected after unilateral stroke.

Recently, Kantak and colleagues used a virtual reality environment to test how patients with stroke coordinate their two arms when performing symmetric and asymmetric reaching movements to move a common, compared to two independent, virtual objects (Kantak et al., 2016). During the independent goal condition, each hand moved its own virtual object to the target position while in the common goal condition the virtual object was moved by an unweighted average of each arm movement. Thus, in the common goal condition there was greater redundancy such that the patients could move the common virtual object with differing contributions from each arm. Similar to the previous studies, we found that during independent-goal condition, patients coupled their arms and contributed almost equally to accomplish their respective goals. In contrast, during common-goal movements, the nonparetic arm contributed significantly greater than the paretic arm and showed poorer temporal coordination in patients compared to bimanual movements during independent goal conditions. Thus despite similar motor requirements, the perceptual nature of the task goals (i.e. common vs. independent) affected coordination between arms in the stroke group. This study was important because it demonstrated that, in addition to action systems, perceptual demands of the task influence bimanual coordination in patients with stroke.

3.2 Kinetic studies of bimanual coordination in post-stroke patients

The ability to generate and control force is necessary for purposeful movements. Therefore, understanding how the two hands coordinate when required to control force has important implications on tasks of daily living. While force generation is a prerequisite for movements, different mechanisms likely underlie the kinematic and kinetic control of bimanual movements. For example, Carson demonstrated that during bimanual isometric force production task, frequency-dependent switching from the anti-phase movements to in-phase movements was not as robust as observed during rhythmic kinematic movements in neurologically-intact individuals (Carson, 1995). Therefore, determining the control of isometric force coordination between arms represents an additional challenge to a comprehensive understanding of bimanual control.

During a common-goal bimanual isometric force production task, patients with stroke produced asymmetric forces similar to kinematic common-goal tasks- i.e. the paretic arm contributed significantly less to the common goal compared to the non-paretic arm (Kang & Cauraugh, 2015). However, this asymmetry was dependent upon the muscle groups being used to generate the force. During wrist and finger extension tasks, force asymmetry was significantly greater because of lesser force contribution from the paretic arm. In contrast, the same patients showed more symmetric movements during gripping tasks. The authors suggested that the spasticity and/or mechanical insufficiency of the wrist and finger flexors likely contribute toward the total force of the paretic grip, therefore reducing the asymmetry between hands during the bimanual gripping task. Thus, force control in bimanual tasks is complex and may rely on the individual features of the task, the muscle groups engaged and constraints of the neuromechanical system of the individual. In addition to making asymmetric movements, stroke patients also show greater inaccuracy and variability during bimanual isometric tasks compared to age-matched controls. Finally, they observed that the non-paretic arm lagged behind the paretic arm, likely because the non-paretic arm was using concurrent feedback to compensate for the inaccuracy and variability of the paretic arm force. This study further highlights that during common-goal bimanual force production tasks, patients may rely more on feedback-based processes than feed-forward processes, thus leading to slower inefficient movements. Similar deficits in cooperative bimanual force tasks have been reported by other investigators (Kang & Cauraugh, 2014, 2015; Lodha, Coombes, & Cauraugh, 2012).

3.3 Bimanual coordination during real-world tasks

Despite their functional relevance, bimanual coordination deficits are almost never quantified during standardized clinical assessments. The assessment scales that include bimanual tasks either quantify speed of bimanual movements (Jebsen’s hand function test or Purdue Peg board test), difficulty of performance (ABILHAND) or independence in performance (Chedoke Arm and Hand Activity Inventory). Almost none of the clinical tests quantify or characterize spatiotemporal coordination between hands that is crucial for optimal function. Given that coordination between arms imparts efficiency for actions of daily living, it is important that clinical outcome measures are developed and tested to characterize bimanual coordination.

Kinematic investigations of real-world tasks have indicated that clinical characteristics of the patients, nature of the tasks and the measures used to quantify coordination influence the findings (Kantak, Zahedi, & McGrath, 2016; McCombe Waller et al., 2006; Wu et al., 2009). For example, reaching to pick up a box with two hands involves the transport phase that is commonly accomplished by the two hands transported to the box in a symmetric-parallel manner. This is followed by the pickup phase where the two hands work in a symmetric cooperative coordination to grip and lift the box. Recently, we demonstrated that patients with stroke retain the ability for bimanual symmetric parallel coordination required during transport of the hands to the box (Kantak et al., 2016). However, upon reaching the box, patients showed significant delays in grasping and picking up the box compared to age-matched controls (longer pickup times), even when normalized to the total movement times. During pick-up of the box, the two hands are required to cooperatively interact with each other to apply perpendicular “grip forces” to secure the box between hands before applying a tangential “load force” to counter gravity and lift the box. These sophisticated cooperative coordination mechanisms may be impaired after stroke leading to delays in pickup times. To further investigate this, we instrumented the box with 3D force transducers to record the grip and load forces during the box pickup. Figure 2 compares the grip and load forces of the two arms between a representative participant with stroke and an age-matched control participant as each grasps and pick up the box. While the control participant coordinates the grip and load forces between the two hands indicated by lesser peaks in grip and load force rates before the box pickup. In contrast, the patient with stroke shows multiple, nonaligned (less coordinated) peaks in the grip and load force rate profiles. These kinematic and kinetic data suggest a deficit in cooperative coordination between the two hands during a naturalistic task. Thus, while symmetric parallel movements (e.g., reaching to the box) are often well-coupled, coordination of symmetric cooperative movements is significantly impaired in patients with stroke.

Fig.2

Kinematic and Kinetic profiles of a neurologically-intact individual and an individual with unilateral stroke. The top row illustrates tangential velocity of each hand. The dotted vertical black line represents the time at which the two hands are transported to contact the box. The vertical solid line represents the initiation of box pick up. Force coordination between the two hands during pick up can best be studied by observing the coordination between the grip force rates (3rd row) and load force rates (5th row) of the two hands. As can be seen in the force rate profiles, the force rates of the right and left hand in the control participant rise together in synchrony for both grip and load forces, indicating an efficient cooperative coordination between the two arms. In contrast, the force rates for each arm in the individual with unilateral subcortical stroke shows multiple peaks between the contact (dotted line) and lift-onset (solid line). Further, the force-rate changes of each hand are not synchronous indicating an impaired cooperative coordination.

Asymmetric actions that form an important part of our daily repertoire are harder and engage a different neural network compared to symmetric actions. Fewer studies have investigated asymmetric naturalistic tasks after stroke. Kantak and colleagues have investigated a “open-the drawer to press the button” task, during which controls move their two hands in opposite directions almost simultaneously- while one hand pulls the drawer, the other hand almost simultaneously reaches forward to press the button. In contrast, patients execute the task in a sequential manner such that the button-pressing paretic hand begins its movement only after the drawer is completely opened by the opening nonparetic hand (Fig. 3). This leads to longer time-lag between the two hands in patients compared to the control participants (Kantak et al., 2016). To further determine if indeed, patients treat the two components of this task (opening the drawer and pressing the button) as two distinct tasks and execute them separately, we compared the performance of the paretic pressing hand under unimanual and bimanual conditions. Specifically, we calculated bimanual cost as the difference in the movement time of the paretic arm between unimanual and bimanual conditions. Coordinating the two arms requires the pressing arm to slow down to ensure that the drawer is opened for the button to be accessible, thus yielding a greater bimanual cost (i.e., longer MT during bimanual compared to unimanual condition). If the two components of the task are planned and executed separately and sequentially, the press hand movement is likely executed similar to a unimanual condition and the difference in the movement time of the pressing hand between unimanual and bimanual conditions would be smaller (lower bimanual cost). In our study, we observed that controls had a significantly greater bimanual cost compared to the stroke group. That is, controls slowed their pressing arm to allow the opening hand to move the drawer open, thus coordinating the two arms. Of importance, we found a significant negative correlation between bimanual cost and the time-lag between the two hands for patients with stroke (Fig. 4). In other words, those patients with longer time lags between hands (poor coordination) had lower bimanual costs. This finding provides greater support to the notion that patients with stroke have difficulty coordinating the two arms during asymmetric tasks and may execute each arm movement as a separate sequential component rather than a well-coordinated single task.

Fig.3

Representative hand kinematics during an asymmetric parallel task of pressing a button made accessible by opening a drawer. Control participant opened the drawer with the dominant hand while pressed the button with the nondominant hand. Participant with stroke opened the drawer with the nonparetic arm and pressed the button with the paretic arm. The reach peak indicates the velocity peak associated with the reach of the hand to the drawer to open it. The pull peak is associated with opening the drawer. The kinematics of control participant show that as he pulls the drawer with his dominant hand, he simultaneously reaches his nondominant hand forward to press the button. This simultaneity of hand movements in opposite direction exemplifies an efficient asymmetric parallel coordination. In contrast, patient with unilateral stroke reaches forward with the paretic arm to press the button only after the nonparetic arm has completely opened the drawer. This dissociated strategy led to longer time-lags between hands.

Fig.4

Inverse significant relationship between bimanual coordination (i.e. Time lag between hands) and bimanual cost (i.e. the MT cost incurred by the paretic press hand during bimanual condition compared to unimanual condition) during the open the drawer to press the button task. Higher bimanual cost indicates longer MTs during bimanual compared to unimanual condition, while negative bimanual cost indicates shorter MTs during bimanual compared to unimanual condition. Shorter time lags between hands (X axis) were associated with greater bimanual cost (Y axis), indicating that patients with good coordination between arms slowed their press hand, perhaps to accommodate the open hand movement, indicating a fine-tuned interaction between the two hands. In contrast those who had longer time lags performed with lesser or even negative bimanual cost. The speeding of the paretic hand in uncoordinated patients (i.e. longer time lags) may reflect a strategy to compensate for delays due to impaired coordination.

In summary, while deficits in bimanual coordination are observed in patients post-stroke, these deficits are less likely to be global and more likely to be dependent on the nature of the task as well as patient characteristics. Both laboratory-based tasks and naturalistic task performance in patients with stroke indicate that symmetric independent-goal and parallel movements that are thought to rely on interlimb coupling are relatively preserved after stroke. However, tasks that require cooperative interaction between arms and asymmetric coordination are likely to be more affected after stroke. More research is warranted to identify how patient characteristics (e.g. lesion location, severity of impairments) interact with task characteristics to yield coordination between the two arms in patients with stroke. This line of research will help identify critical factors that impair bimanual coordination after stroke.

3.4 Relationship of bimanual coordination with unimanual impairments/capability

While it is tempting to attribute the impaired bimanual performance to the unimanual deficits in the paretic arm, the reduction in bilateral arm activity is only weakly associated with the unimanual capability of the weaker arm (tested using the Action Research Arm test). Further, studies have consistently failed to report strong relationships between motor impairments or performance of the paretic arm and bimanual coordination. Importantly, the relationship between paretic arm impairments and bimanual coordination is strongly influenced by patient and task characteristics. For example, cooperative coordination during pickup of the box did not strongly correlate with the motor impairments such as Fugl-Meyer score or strength. Instead, impaired cooperative coordination was strongly associated with greater proprioceptive deficits of the weaker arm (Kantak et al., 2016). Thus, when the hands need to interact with each other for an effective pickup, they likely rely on sensory systems for predictive and feedback-based processes. In addition to the paretic arm impairments, multiple studies have identified sensorimotor deficits in the ipsilesional, nonparetic arm that may significantly contribute to difficulty coordinating the two arms. While the impact of the ipsilesional arm on bimanual coordination has not been systematically investigated, it is highly plausible that deficits in both ipsilesional and contralesional arms may impair bimanual coordination. More research is needed to determine the relative contributions of ipsilesional and contralesional arm deficits to bimanual coordination across different types of task. Further, it remains to be determined if recovering independent unimanual control of the ipsilesional and contralesional arm through unimanual training could improve interlimb coordination during different classes of bimanual actions.

Unilateral stroke may also disrupt specialized brain regions and/or neural networks that are critical to bimanual coordination. For example, there is evidence that right superior temporal gyrus activity is necessary for bimanual coordination of a common-goal bimanual task (Julie Duque et al., 2010). A stroke-induced lesion in this region may be associated with impaired coordination during functional actions relying on the specialized function of the right STG. Further, specific temporal and spatial interrelationship between arm actions relies on a strictly-timed interaction between brain regions from the two hemispheres through specific neural networks. Following stroke there is evidence for altered interhemispheric interactions that impair unimanual performance of the weaker arm (Duque et al., 2005; Murase, Duque, Mazzocchio, & Cohen, 2004). It is conceivable that such deficient interhemispheric interactions may disrupt the necessary neural communication for coordinated bimanual actions. Finally, a large body of research in healthy adults indicates that perceptual factors play a crucial role in bimanual coordination (Mechsner et al., 2001; Shea et al., 2016). For example, while picking a tray, we use perceptual cues related to the amount of weight and it distribution along the tray to apply well-coordinated forces between the two hands for a smooth pickup. It is not known how perceptual impairments in patients with stroke influences the coordination between the two arms during actions. Given the complexity and interaction of contributing factors to bimanual coordination, there is a dire need to focus research efforts on understanding and improving coordinated bimanual function after stroke. Identifying the factors associated with poor bimanual coordination may help guide rehabilitation interventions to improve skilled coordinated bimanual arm use for daily function. As outlined in the subsequent section, current rehabilitation strategies fall short of specifically targeting bimanual coordination, thus dampening a possibility of promoting full functional + recovery.

4 What is missing in current rehabilitation strategies?

The long-term goal of post-stroke arm rehabilitation after stroke is to (re)integrate the paretic arm in daily functional actions such that functional tasks are performed with greater skill and efficiency. Since unilateral stroke leads to the most obvious deficits in the contralesional arm, most rehabilitation research has predominantly focused on remediating impairments (e.g. robot-based therapies (Kantak, Jones-Lush, Narayanan, Judkins, & Wittenberg, 2013; Lo et al., 2010; Mazzoleni et al., 2013; Rickards et al., 2012) and improving unimanual performance of the contralesional weaker arm (e.g. constraint-induced movement therapy (Rickards et al., 2012; Wolf et al., 2006, 2010; Wu et al., 2007). Very few of these studies have tested the effects of unimanual paretic arm improvements on bimanual function and/or coordination; thus making it difficult to determine if indeed skill performance transfers from unimanual to bimanual conditions.

Studies in healthy able-bodied adults however have challenged the implicit assumption of transfer from unimanual to bimanual performance (Hinder, Carroll, & Summers, 2013; Nozaki, Kurtzer, & Scott, 2006). Nozaki and colleagues trained subjects to perform unimanual and bimanual reaching while a force-field was applied to one of their arms. Subjects demonstrated partial, but not complete transfer of learning within the same limb between unimanual and bimanual conditions. Further, this transfer occurred only when there was an appropriate contextual cue indicating the context similar to training. Therefore, it is important to recognize that transfer from unimanual to bimanual task conditions may require complex cognitive and perceptual strategies that likely require targeted training (Gordon, 2011; Mullick, Subramanian, & Levin, 2015). Such systematic studies have not yet been done in patients with stroke. More research is needed to determine the factors that influence the transfer of learning after unimanual practice to bimanual coordination and/or performance, particularly in people with stroke.

Besides unimanual approaches, two major classes of bilateral therapeutic approaches evident in the literature have also had limited impact on interactive coordination between arms. One approach relies on improving paretic arm movement through symmetric bimanual “coupling”, operationally administered as simultaneous (and often rhythmic) use of homologous muscles in each arm (Arya & Pandian, 2014; Cauraugh, 2004; Cauraugh, Lodha, Naik, & Summers, 2010; Cunningham et al., 2002; Harris-Love et al., 2005; Stinear et al., 2008; Stoykov, Lewis, & Corcos, 2009; Stoykov & Stinear, 2010; Whitall et al., 2000a). Symmetric coupling likely relies on inherent neurophysiological and mechanical linkages that constrain the two arms to act as a unit, thus facilitating the activation of the weaker arm (Cunningham, Stoykov, & Walter, 2002; Rose & Winstein, 2004). Repetitive coupled bimanual movements have been used as priming to “ready” the neuromuscular system for a more plastic response to subsequent training/practice. Alternately, some studies have used repetitive coupled bimanual movements as training without any subsequent task practice or therapy. When symmetric coupled movements are repetitively and rhythmically administered as priming before task practice, studies have shown beneficial effects on unimanual function of the paretic arm (Stinear et al., 2008). Repetitive rhythmic symmetric bilateral priming is shown to reduce the transcallosal inhibition from the contralesional to ipsilesional motor cortex and increase the ipsilesional motor cortex excitability, thereby enhancing unimanual performance of the paretic arm (Byblow et al., 2012). Surprisingly, while the basis of the bilateral priming approach was theoretically based on interaction between the two arms, the focus of outcomes in these studies has been on unimanual performance changes. No studies have surprisingly assessed the effects of priming on bimanual coordination performance or training.

Some studies have used repetitive bilateral (often symmetric) movements as training. This training may operationally take similar form as bimanual priming (i.e. rhythmic bimanual movements); however unlike priming, training is not followed by task practice/therapy. A Cochrane review indicated that bilateral training may be no more (or less) effective than usual care or other upper limb interventions for performance in ADL, functional movements of the upper limb or motor impairment outcomes (Coupar, Pollock, van Wijck, Morris, & Langhorne, 2010). One limitation is that the precise dosage, intensity and timing of coupled bilateral training have not yet been distilled to optimize benefits for unimanual and/or bimanual function. DeJong and Lang found that beneficial effects of the bilateral condition on paretic arm performance were evident only during a maximal force production condition (Dejong & Lang, 2012). Most coupled bilateral training protocols do not provide high enough intensity and/or challenge to yield greater benefits compared to other approaches (Corti, McGuirk, Wu, & Patten, 2012), thus dampening the benefits of training. Another important limitation of coupled bilateral training is that they disregard the fact that arms engage in an infinite variety of coordinated behaviors that involve different actions of each arm such as washing dishes, moving a heavy teapot with one hand while carrying a small cup in the other. Such coordinated and interactive control of two arms requires overriding the coupled inherent linkages through a complex goal-directed interaction between motor and cognitive-perceptual systems of the nervous system (Mullick et al., 2015; Sleimen-Malkoun et al., 2011). It seems crucial that bilateral approaches train coordination modes beyond rhythmic symmetrical movements to have an impact on daily function.

The other bilateral approach focuses on task practice of bimanual actions with the goal of improving movement pattern and speed of the paretic arm in the context of real-world bimanual tasks. These approaches do not specifically target the interactive spatiotemporal coordination between hands during these tasks. For that reason, real-world bimanual task practice has had a limited impact of upper extremity performance. Further, very few studies have assessed performance on bimanual tasks following training, with almost none specifically testing inter-limb coordination during bimanual tasks.

Pertinent to the current paper, a philosophical question arises that has important practical implications – what greater advantage do coordinated bimanual actions offer in daily actions compared to uncoordinated movements? Given the redundancy of degrees-of-freedom with two arms, any task may be accomplished through different combinations of movements of each arm. For example, in order to retrieve a pen from a drawer, one may use one hand to open the drawer and once opened, use the same hand or the other hand to retrieve the pen (serial execution). Alternately, as is observed in most able-bodied adults, while one hand pulls the drawer open, the other hand almost simultaneously reaches forward to grasp the pen (coordinated execution). The simultaneity of the hand movements likely make execution faster and smoother than serial movements of the two hands. That is, when multiple sub-goals need to be accomplished for a specific task completion, simultaneous coordinated movements of the two arms to accomplish each of those sub-goals is likely to be a more efficient solution to the multiple degrees-of-freedom problem. Therefore, coordination between the two arms is critical for economical and efficient movements. This concept has particular significance in stroke rehabilitation. If coordination between the two arms is impaired, the efficiency imparted by simultaneous use for certain bimanual actions may no longer be available. Patients may then prefer to accomplish the whole movement with their faster nonparetic arm. For example, in order to retrieve the pen from the drawer, patients may use the nonparetic arm to open the drawer and then grasp the pen with the same hand once the drawer is opened. Thus, impaired coordination may nullify the efficiency of using two hands and lead to increased reliance on the nonparetic arm during daily actions. In an observational study (Haaland et al., 2012), Haaland and colleagues examined the relationship between different patterns of arm use and everyday functional performance in a cohort of stroke patients with moderate impairments (mean Fugl-Meyer score- 46.5). Patterns of arm use were recorded using accelerometers on each wrist; and functional performance was measured using instrumented ADLs from the Functional Impact Assessment. They observed that greater simultaneous use of both arms was associated with better scores on the functional instrumented ADLs. These findings indicate that patient’s ability for coordinated use of both arms together is likely a predictor of real-world arm function after stroke. Therefore, it is crucial that coordination between the two arms is assessed and trained to improve functional performance in patients with stroke. Surprising, there is a dearth of research investigating the quantification, understanding the mechanisms and treatment of bimanual coordination (Obhi, 2004), particularly in people with stroke. In the next section, we outline important research directions to advance the scientific and clinical bases for incorporation of bimanual coordination in arm rehabilitation.

5 Future directions

Training coordination between arms for different types of bimanual actions needs to be an important component of arm rehabilitation to achieve more complete functional recovery. To reap maximum benefits of bimanual coordination training, it is imperative to quantify performance deficits specific to coordination between arms, determine the nature of tasks that show such deficits, identify patient characteristics and neural mechanisms associated with specific bimanual coordination deficits; and invent and test new behavioral and neurophysiological strategies to improve bimanual coordination between arms. To realize this goal, research efforts need to target different levels of the translational research pipeline and integrate the emergent findings to inform clinical decision-making in arm rehabilitation.

While research efforts so far have been focused on “coupled” bimanual movements than cooperative bimanual actions, we suggest that future research expand its scope to include different types of bimanual actions evident in daily life. Broadly, we propose that determining the behavioral and neural mechanisms that underlie normal and disordered coordination of bimanual action will inspire novel, scientifically-sound behavioral and neurophysiological approaches to improve bimanual coordination and spontaneous use of the weaker arm in daily actions.

Determine post-stroke deficits in bimanual coordination: Research characterizing deficits in bimanual coordination that extends beyond symmetric or asymmetric independent goals (i.e. coupling) will provide a more comprehensive understanding of which coordination “modes” are impaired in patients with stroke. Using laboratory-based tasks and real-world tasks with instrumented objects may provide rich kinematic and kinetic data (e.g. Fig. 2) to identify quantify and characterize bimanual coordination deficits.

Determining time-course of bimanual coordination improvements after stroke: While improvements in bimanual coordination are evident early after stroke (Metrot et al., 2013), it is not clear if the time-course of bimanual coordination improvements follows a similar trajectory as that of the paretic and non-paretic arms. Further, the role of learned nonuse is unclear. While on one hand learned nonuse of the paretic arm may limit its use in bimanual tasks, it is also plausible that task requirements may afford greater engagement of the paretic arm. Animal studies have shown that incorporating bimanual activities earlier after stroke prevents the maladaptive neuroplastic changes associated with paretic arm nonuse (Kerr, Wolke, Bell, & Jones, 2013). With advances in the quantification of learned arm-use (Han et al., 2013), these competitive hypotheses may be carefully tested.

Determine contributing factors to deficient coordination: In rehabilitation, not one size fits all (Mayo et al., 2016; Shah-Basak et al., 2015). Identifying patient characteristics that are associated with deficient bimanual coordination will allow selection of patients for targeted interventions to improve bimanual coordination. Particularly, quantifying the variance in bimanual coordination during different actions that can be attributed to unimanual impairments and capability of the paretic and nonparetic arms may allow focusing interventions in rehabilitation. Further, determining how connectivity between the key areas of two brain hemispheres that is shown to be impaired following stroke influence the ability for bimanual movements will allow prediction of deficits based on neurophysiological data (de Vries, Daffertshofer, Stegeman, & Boonstra, 2016; Fujiyama, Van Soom, Rens, Cuypers, et al., 2016; Fujiyama, Van Soom, Rens, Gooijers, et al., 2016; Loehrer et al., 2016). Identifying the brain-behavior relationship for bimanual coordination through brain imaging will also afford new neurophysiological ways to improve bimanual coordination.

Perceptual influences in bimanual coordination after stroke: Psychophysical research has demonstrated that perceptual system has a strong influence on motor performance. For example, providing simple unified feedback for both hands rather than separate feedback about each hand allowed individuals to perform highly complex bimanual movements that were deemed almost impossible (Diedrichsen, 2007; Diedrichsen et al., 2006b; Kantak et al., 2016; Shea et al., 2016). Thus, systematically investigating the largely-ignored perceptual-motor interactions after stroke may provide novel strategies to influence bimanual motor capabilities in patients with stroke.

Training bimanual coordination:

Identifying optimal dosage, intensity and mode of training: For coupled symmetric training, the focus of training has been on the entrainment using auditory cues (Whitall, McCombe Waller, Silver, & Macko, 2000b) and/or passive-to-active progression (Stinear et al., 2008). The success of these approaches compared to other interventions has been limited (van Delden et al., 2009; van Delden, Beek, Roerdink, Kwakkel, & Peper, 2014). These “coupled training” approaches so far have not required maximum effort from the paretic and/or nonparetic arms. Recent studies have provided sufficient evidence to indicate that intensity (or effort) of task practice is critical in driving neuroplastic and behavioral recovery (Hornby et al., 2011; Hornby et al., 2016). Given that the benefits of bimanual condition may be reaped more strongly during maximal effort (Dejong & Lang, 2012), the lack of efficacy may be in part attributed to inadequate dosage and intensity of bimanual training. Research efforts are strongly indicated to glean the parameters (e.g. intensity, dose) of bilateral coupled training programs.

Perceptual cues in training: Building upon the perceptual motor interactions after stroke outlined before, determining perceptual cues that may be used in training to improve bimanual coordination in long-term. Virtual reality that has begun to impact rehabilitation research may provide uniquely designed environments and feedback to optimally train coordination between arms for simulated bimanual tasks (Lee, Kim, & Lee, 2016; Subramanian, Lourenço, Chilingaryan, Sveistrup, & Levin, 2013). Such innovative and collaborative solutions may not only enable complex feedback presentations, but allow testing and training at home to encourage practice and subsequent transfer to real-world.

Training and transfer between unimanual and bimanual contexts: Unimanual performance improvements have been reported following unilateral (Lang et al., 2016; Wolf et al., 2010) and bilateral “coupled” training (Cauraugh, Naik, Lodha, Coombes, & Summers, 2011; Stinear, Petoe, Anwar, Barber, & Byblow, 2013; Whitall et al., 2000b). Determining the extent of transfer of unimanual improvements to classes of coordinated bimanual actions will provide bases for selecting and designing interventions to improve specific bimanual task performance. Similarly, determining if practice of the paretic arm in the context of a coordinated bimanual task transfers to improved unimanual capability of the paretic arm. Such basic science questions have critical translational significance.

Noninvasive brain stimulation combined with bimanual practice: It is not a surprise that unimanual and bimanual interventions implement their respective benefits through distinct neural substrates and neurophysiologic networks (Whitall et al., 2011). Non-invasive brain stimulation protocols have demonstrated relatively consistent beneficial effects on unimanual performance (Kang, Summers, & Cauraugh, 2016; Tretriluxana, Kantak, Tretriluxana, Wu, & Fisher, 2013). However, NIBS protocol that improved unimanual performance impaired bimanual performance (Takeuchi, Tada, Matsuo, & Ikoma, 2012). These findings suggest that tailoring the brain stimulation protocols to the neurophysiological substrates of bimanual interventions may yield better bimanual coordination outcomes. Particularly, determining and targeting the neural substrates for cooperative bimanual actions may provide benefits that likely transfer to real-world use.

Determining who benefits from bimanual training: It is being increasingly acknowledged that “one size doesn’t fit all” in stroke rehabilitation and individual patient differences play a significant role in determining the benefits reaped from rehabilitation interventions. Therefore, characterizing the impairments, lesion location, residual connectivity and psychosocial factors may allow identification of patients that may optimally benefit from bimanual priming and/or training approaches.

6 Conclusions

Coordinated use of both upper extremities is an important component of skilled arm function that characterizes most daily activities. Despite its ubiquitous nature and functional significance, bimanual coordination has not been well-studied in stroke rehabilitation literature. Current heavy focus on unimanual deficits, difficulty in characterization and quantification of bimanual coordination for different tasks and lack of knowledge about the best ways to treat deficits in bimanual coordination are among biggest obstacles to systematic study of bimanual coordination post-stroke. Based on our current understanding of the scientific bases of interlimb coordination, we provide a rather simplistic taxonomy to categorize different forms of coordination. While the current literature in stroke rehabilitation has focused on symmetric coupling between arms, the present perspective calls for a more comprehensive study of bimanual coordination that includes other forms of coordination. Future research is indicated to characterize specific bimanual coordination deficits, understand the mechanisms implementing these deficits and finding optimal ways to improve bimanual coordination is needed. Incorporating strategies to improve coordinated and skill use of both arms is likely an important component of arm rehabilitation that needs more research for optimal implementation.

References

Aramaki

, Osu

, & Sadato

(2010). Resource-demanding versus cost-effective bimanual interaction in the brain. Experimental Brain Research, 203(2), 407–418.

Arya

K.N.

, & Pandian

(2014). Interlimb neural coupling: Implications for poststroke hemiparesis. Annals of Physical and Rehabilitation Medicine, 57, 696–713.

Bailey

R.R.

, Birkenmeier

R.L.

, & Lang

C.E.

(2015). Real-world affected upper limb activity in chronic stroke: An examination of potential modifying factors. Topics in Stroke Rehabilitation, 22(1), 26–33.

Bailey

R.R.

, Klaesner

J.W.

, & Lang

C.E.

(2014). Anaccelerometry-based methodology for assessment of real-worldbilateral upper extremity activity. PLoS One, 9(7), e103135.

Bailey

R.R.

, Klaesner

J.W.

, & Lang

C.E.

(2015a). Quantifying real-world upper-limb activity in nondisabled adults and adults with chronic stroke. Neurorehabilitation and Neural Repair, 29(10), 969–978.

Bailey

R.R.

, Klaesner

J.W.

, & Lang

C.E.

(2015b). Quantifying real-world upper-limb activity in nondisabled adults and adults with chronic stroke. Neurorehabilitation and Neural Repair, 29(10), 969–978.

Beer

R.F.

, Dewald

J.P.

, & Rymer

W.Z.

(2000). Deficits in the coordination of multijoint arm movements in patients with hemiparesis: Evidence for disturbed control of limb dynamics. Experimental Brain Research, 131(3), 305–319.

Blinch

, Cameron

B.D.

, Franks

I.M.

, Carpenter

M.G.

, & Chua

(2015). Facilitation and interference during the preparation of bimanual movements: Contributions from starting locations, movement amplitudes, and target locations. Psychological Research, 79(6), 978–988.

Blinch

, Cameron

B.D.

, Franks

I.M.

, & Chua

(2011). Bimanual reaches with symbolic cues exhibit errors in target selection. Experimental Brain Research, 212(4), 541–554.

10.

Byblow

W.D.

, Stinear

C.M.

, Smith

M.-C.

, Bjerre

, Flaskager

B.K.

, & McCambridge

A.B.

(2012). Mirror symmetric bimanualmovement priming can increase corticomotor excitability andenhance motor learning. PLoS One, 7(3), e33882.

11.

Carson

R.G.

(1995). The dynamics of isometric bimanual coordination. Experimental Brain Research, 105(3), 465–476.

12.

Cauraugh

J.H.

(2004). Coupled rehabilitation protocols and neural plasticity: Upper extremity improvements in chronic hemiparesis. Restorative Neurology and Neuroscience, 22(3-5), 337–347.

13.

Cauraugh

J.H.

, Lodha

, Naik

S.K.

, & Summers

J.J.

(2010). Bilateral movement training and stroke motor recovery progress: A structured review and meta-analysis. Human Movement Science, 29(5), 853–870.

14.

Cauraugh

J.H.

, Naik

S.K.

, Lodha

, Coombes

S.A.

, & Summers

J.J.

(2011). Long-term rehabilitation for chronic stroke arm movements: A randomized controlled trial. Clinical Rehabilitation, 25(12), 1086–1096.

15.

Cohn

, Bender

M.B.

, Bender

M.B.

, T.H.L, Cohn

, Babinski

, Akelaitis

A.J.

, Nielsen

J.M.

, R.R.B. (1951). Interaction in bilaterally simultaneous voluntary motor function. Archives of Neurology And Psychiatry, 65(4), 472.

16.

Conroy

S.S.

, Whitall

, Dipietro

, Jones-Lush

L.M.

, Zhan

, Finley

M.A.

, & … Bever

C.T.

(2011). Effect of gravity on robot-assisted motor training after chronic stroke: A randomized trial. Archives of Physical Medicine and Rehabilitation, 92(11), 1754–1761.

17.

Corti

, McGuirk

T.E.

, Wu

S.S.

, & Patten

(2012). Differential effects of power training versus functional task practice on compensation and restoration of arm function after stroke. Neurorehabilitation and Neural Repair, 26(7), 842–854.

18.

Coupar

, Pollock

, van Wijck

, Morris

, & Langhorne

(2010). Simultaneous bilateral training for improving arm function after stroke. The Cochrane Database of Systematic Reviews, (4), CD006432.

19.

Cunningham

C.L.

, Stoykov

M.E.P.

, & Walter

C.B.

(2002). Bilateral facilitation of motor control in chronic hemiplegia. Acta Psychologica, 110(2-3), 321–337.

20.

d’Avella

, & Lacquaniti

(2013). Control of reaching movements by muscle synergy combinations. Frontiers in Computational Neuroscience, 7, 42.

21.

Daly

J.J.

, Hrovat

, Holcomb

, & Pundik

(2014). Brain control of functional reach in healthy adults and stroke survivors. Restorative Neurology and Neuroscience, 32(5), 559–573.

22.

de Vries

I.E.J.

, Daffertshofer

, Stegeman

D.F.

, & Boonstra

T.W.

(2016). Functional connectivity in the neuromuscular system underlying bimanual coordination. Journal of Neurophysiology, 116(6), 2576–2585.

23.

Dejong

S.L.

, & Lang

C.E.

(2012). Comparison of unilateral versus bilateral upper extremity task performance after stroke. Topics in Stroke Rehabilitation, 19(4), 294–305.

24.

Diedrichsen

(2007). Optimal task-dependent changes of bimanual feedback control and adaptation. Current Biology: CB, 17(19), 1675–1679.

25.

Diedrichsen

, Grafton

, Albert

, Hazeltine

, & Ivry

R.B.

(2006a). Goal-selection and movement-related conflict duringbimanual reaching movements. Cerebral Cortex (New York, N.Y.: 1991), 16(12), 1729–1738.

26.

Diedrichsen

, Grafton

, Albert

, Hazeltine

, & Ivry

R.B.

(2006b). Goal-selection and movement-related conflict duringbimanual reaching movements. Cerebral Cortex (New York, N.Y.: 1991), 16(12), 1729–1738.

27.

Diedrichsen

, & Gush

(2009). Reversal of bimanual feedback responses with changes in task goal. Journal of Neurophysiology, 101(1), 283–288.

28.

Diedrichsen

, Shadmehr

, & Ivry

R.B.

(2010). The coordination of movement: Optimal feedback control and beyond. Trends in Cognitive Sciences, 14(1), 31–39.

29.

Duque

, Davare

, Delaunay

, Jacob

, Saur

, Hummel

, & … Olivier

(2010). Monitoring coordination during bimanual movements: Where is the mastermind? Journal of Cognitive Neuroscience, 22(3), 526–542.

30.

Duque

, Mazzocchio

, Dambrosia

, Murase

, Olivier

, & Cohen

L.G.

(2005). Kinematically specific interhemispheric inhibition operating in the process of generation of a voluntary movement. Cerebral Cortex (New York, N.Y.: 1991), 15(5), 588–593.

31.

Faria-Fortini

, Michaelsen

S.M.

, Cassiano

J.G.

, & Teixeira-Salmela

L.F.

(2011). Upper extremity function in stroke subjects: Relationships between the international classification of functioning, disability, and health domains. Journal of Hand Therapy: Official Journal of the American Society of Hand Therapists, 24(3), 257–264; quiz 265.

32.

Franz

E.A.

(2003). Bimanual Action Representation: A Window on Human Evolution, 259–288.

33.

Fujiyama

, Van Soom

, Rens

, Cuypers

, Heise

K.-F.

, Levin

, & Swinnen

S.P.

(2016). Performing two different actions simultaneously: The critical role of interhemispheric interactions during the preparation of bimanual movement. Cortex; a Journal Devoted to the Study of the Nervous System and Behavior, 77, 141–154.

34.

Fujiyama

, Van Soom

, Rens

, Gooijers

, Leunissen

, Levin

, & Swinnen

S.P.

(2016). Age-related changes in frontal network structural and functional connectivity in relation to bimanual movement control. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 36(6), 1808–1822.

35.

George Hornby

, Straube

D.S.

, Kinnaird

C.R.

, Holleran

C.L.

, Echauz

A.J.

, Rodriguez

K.S.

, & … Narducci

E.A.

(2011). Importance of specificity, amount, and intensity of locomotor training to improve ambulatory function in patients poststroke. Topics in Stroke Rehabilitation, 18(4), 293–307.

36.

Gerloff

, & Andres

F.G.

(2002). Bimanual coordination and interhemispheric interaction. Acta Psychologica, 110(2-3), 161–186.

37.

Gooijers

, & Swinnen

S.P.

(2014). Interactions between brain structure and behor: The corpus callosum and bimanual coordination. Neuroscience and Biobehavioral Reviews, 43, 1–19.

38.

Gordon

A.M.

(2011). To constrain or not to constrain, and other stories of intensive upper extremity training for children with unilateral cerebral palsy. Developmental Medicine and Child Neurology, 53(SUPPL.4), 56–61.

39.

Haaland

K.Y.

, Mutha

P.K.

, Rinehart

J.K.

, Daniels

, Cushnyr

, & Adair

J.C.

(2012). Relationship between arm usage and instrumental activities of daily living after unilateral stroke. Archives of Physical Medicine and Rehabilitation, 93(11), 1957–1962.

40.

Han

C.E.

, Kim

, Chen

, Lai

Y.-H.

, Lee

J.-Y.

, Osu

, & … Schweighofer

(2013). Quantifying arm nonuse in individuals poststroke. Neurorehabilitation and Neural Repair, 27(5), 439–447.

41.

Harris-Love

M.L.

, Waller

S.M.

, & Whitall

(2005). Exploiting interlimb coupling to improve paretic arm reaching performance in people with chronic stroke. Archives of Physical Medicine and Rehabilitation, 86(11), 2131–2137.

42.

Hinder

M.R.

, Carroll

T.J.

, & Summers

J.J.

(2013). Transfer of ballistic motor skill between bilateral and unilateral contexts in young and older adults: Neural adaptations and behavioral implications. Journal of Neurophysiology, 109(12), 2963–2971.

43.

Hornby

T.G.

, Holleran

C.L.

, Hennessy

P.W.

, Leddy

A.L.

, Connolly

, Camardo

, & … Roth

E.J.

(2016). Variable intensive early walking poststroke (VIEWS): A randomized controlled trial. Neurorehabilitation and Neural Repair, 30(5), 440–450.

44.

Howard

I.S.

, Ingram

J.N.

, Körding

K.P.

, & Wolpert

D.M.

(2009). Statistics of natural movements are reflected in motor errors. Journal of Neurophysiology, 102(3), 1902–1910.

45.

Jarrassé

, Ribeiro

, Sahbani

, Bachta

, & Roby-Brami

(2014). Analysis of hand synergies in healthy subjects during bimanual manipulation of various objects. Journal of NeuroEngineering and Rehabilitation, 11(1), 113.

46.

Jirsa

V.K.

, Fuchs

, & Kelso

J.A.

(1998). Connecting cortical and behavioral dynamics: Bimanual coordination. Neural Computation, 10(8), 2019–2045.

47.

Kang

, & Cauraugh

J.H.

(2014). Force control improvements in chronic stroke: Bimanual coordination and motor synergy evidence after coupled bimanual movement training. Experimental Brain Research, 232(2), 503–513.

48.

Kang

, & Cauraugh

J.H.

(2015). Force control in chronic stroke. Neuroscience and Biobehavioral Reviews, 52, 38–48.

49.

Kang

, Summers

J.J.

, & Cauraugh

J.H.

(2016). Transcranial direct current stimulation facilitates motor learning post-stroke: A systematic review and meta-analysis. Journal of Neurology, Neurosurgery, and Psychiatry, 87(4), 345–355.

50.

Kantak

, McGrath

, & Zahedi

(2016a). Goal conceptualization and symmetry of arm movements affect bimanual coordination in individuals after stroke. Neuroscience Letters, 626, 86–93.

51.

Kantak

S.S.

, Jones-Lush

L.M.

, Narayanan

, Judkins

T.N.

, & Wittenberg

G.F.

(2013). Rapid plasticity of motor corticospinal system with robotic reach training. Neuroscience, 247, 55–64.

52.

Kantak

S.S.

, Zahedi

, & McGrath

R.L.

(2016a). Task-dependent bimanual coordination after stroke: Relationship with sensorimotor impairments. Archives of Physical Medicine and Rehabilitation, 97(5), 798–806.

53.

Kelso

J.S.

(1983). Phase transitions and critical behavior in human bimanual coordination.

54.

Kerr

A.L.

, Wolke

M.L.

, Bell

J.A.

, & Jones

T.A.

(2013). Post-stroke protection from maladaptive effects of learning with the non-paretic forelimb by bimanual home cage experience in C57BL/6 mice. Behavioural Brain Research, 252, 180–187.

55.

Kilbreath

S.L.

, & Heard

R.C.

(2005). Frequency of hand use in healthy older persons. The Australian Journal of Physiotherapy, 51(2), 119–122.

56.

Kwakkel

, Kollen

, & Lindeman

(2004). Understanding the pattern of functional recovery after stroke: Facts and theories. Restorative Neurology and Neuroscience, 22(3-5), 281–299.

57.

Lang

C.E.

, Strube

M.J.

, Bland

M.D.

, Waddell

K.J.

, Cherry-Allen

K.M.

, Nudo

R.J.

, & … Birkenmeier

R.L.

(2016). Dose response of task-specific upper limb training in people at least 6 months poststroke: A phase II, single-blind, randomized, controlled trial. Annals of Neurology, 80(3), 342–354.

58.

Lee

, Kim

, & Lee

B.-H.

(2016). Effect of virtual reality-based bilateral upper extremity training on upper extremity function after stroke: A randomized controlled clinical trial. Occupational Therapy International, 23(4), 357–368.

59.

Lewis

G.N.

, & Byblow

W.D.

(2004). Bimanual coordination dynamics in poststroke hemiparetics. Journal of Motor Behavior, 36(2), 174–188.

60.

A.C.

, Guarino

P.D.

, Richards

L.G.

, Haselkorn

J.K.

, Wittenberg

G.F.

, Federman

D.G.

, & … Peduzzi

(2010). Robot-assisted therapy for long-term upper-limb impairment after stroke. New England Journal of Medicine, 362(19), 1772–1783.

61.

Lodha

, Coombes

S.A.

, & Cauraugh

J.H.

(2012). Bimanual isometric force control: Asymmetry and coordination evidence post stroke. Clinical Neurophysiology: Official Journal of the International Federation of Clinical Neurophysiology, 123(4), 787–795.

62.

Loehrer

P.A.

, Nettersheim

F.S.

, Jung

, Weber

, Huber

, Dembek

T.A.

, & … Timmermann

(2016). Ageing changes effective connectivity of motor networks during bimanual finger coordination. NeuroImage, 143, 325–342.

63.

Long

, Tazoe

, Soteropoulos

D.S.

, & Perez

M.A.

(2016). Interhemispheric connectivity during bimanual isometric force generation. Journal of Neurophysiology, 115(3), 1196–1207.

64.

Lowrey

, Jackson

, Bagg

, Dukelow

, Scott

(2014). A novel robotic task for assessing impairments in bimanual coordination post-stroke. S3, 002.

65.

MacKenzie

, & RG

(1985). Bimanual Coordination. In Advances in Psychology (pp. 345–358).

66.

Maki

, Wong

K.F.K.

, Sugiura

, Ozaki

, & Sadato

(2008). Asymmetric control mechanisms of bimanual coordination: An application of directed connectivity analysis to kinematic and functional MRI data. NeuroImage, 42(4), 1295–1304.

67.

Mayo

N.E.

, Aburub

, Brouillette

M.-J.

, Kuspinar

, Moriello

, Rodriguez

A.M.

, & Scott

(2017). In support of an individualized approach to assessing quality of life: Comparison between Patient Generated Index and standardized measures across four health conditions. Quality of Life Research, 26(3), 601–609.

68.

Mazzoleni

, Sale

, Tiboni

, Franceschini

, Carrozza

M.C.

, & Posteraro

(2013). Upper limb robot-assisted therapy in chronic and subacute stroke patients: A kinematic analysis. e. American Journal of Physical Medicine & Rehabilitation/Association of Academic Physiatrists, 92(10 Suppl 1), 26–37.

69.

McCombe Waller

, Harris-Love

, Liu

, & Whitall

(2006). Temporal coordination of the arms during bilateral simultaneous and sequential movements in patients with chronic hemiparesis. Experimental Brain Research, 168(3), 450–454.

70.

McCombe Waller

, Whitall

, Jenkins

, Magder

L.S.

, Hanley

D.F.

, Goldberg

, & Luft

A.R.

(2014). Sequencing bilateral and unilateral task-oriented training versus task oriented training alone to improve arm function in individuals with chronic stroke. BMC Neurology, 14, 236.

71.

Mechsner

, Kerzel

, Knoblich

, & Prinz

(2001). Perceptual basis of bimanual coordination. Nature, 414(6859), 69–73.

72.

Mechsner

, Knoblich

, & Mechsner

(2004). Do muscles matter for coordinated action? Journal of Experimental Psychology. Human Perception and Performance, 30(3), 490–503.

73.

Michielsen

M.E.

, Selles

R.W.

, Stam

H.J.

, Ribbers

G.M.

, & Bussmann

J.B.

(2012). Quantifying nonuse in chronic stroke patients: A study into paretic, nonparetic, and bimanual upper-limb use in daily life. Archives of Physical Medicine and Rehabilitation, 93(11), 1975–1981.

74.

Mudie

M.H.

, & Matyas

T.A.

(2000). Can simultaneous bilateral movement involve the undamaged hemisphere in reconstruction of neural networks damaged by stroke? Disability and Rehabilitation, 22(1-2), 23–37.

75.

Mullick

A.A.

, Subramanian

S.K.

, & Levin

M.F.

(2015). Emerging evidence of the association between cognitive deficits and arm motor recovery after stroke: A meta-analysis. Restorative Neurology and Neuroscience, 33(3), 389–403.

76.

Murase

, Duque

, Mazzocchio

, & Cohen

L.G.

(2004). Influence of interhemispheric interactions on motor function in chronic stroke. Annals of Neurology, 55(3), 400–409.

77.

Nozaki

, Kurtzer

, & Scott

S.H.

(2006). Limited transfer of learning between unimanual and bimanual skills within the same limb. Nature Neuroscience, 9(11), 1364–1366.

78.

Obhi

S.S.

(2004). Bimanual coordination: An unbalanced field of research. Motor Control, 8(2), 111–120.

79.

Perez

M.A.

, Butler

J.E.

, & Taylor

J.L.

(2014). Modulation of transcallosal inhibition by bilateral activation of agonist and antagonist proximal arm muscles. Journal of Neurophysiology, 111(2), 405–414.

80.

Rand

, & Eng

J.J.

(2010). Arm-hand use in healthy older adults. The American Journal of Occupational Therapy: Official Publication of the American Occupational Therapy Association, 64(6), 877–885.

81.

Rickards

, Taub

, Sterling

, Graham

M.J.

, Barghi

, Uswatte

, & Mark

V.W.

(2012). Brain parenchymal fractionpredicts motor improvement following intensive task-oriented motorrehabilitation for chronic stroke. Restorative Neurology and Neuroscience, 30(5), 355–361.

82.

Roh

, Rymer

W.Z.

, & Beer

R.F.

(2015). Evidence for altered upper extremity muscle synergies in chronic stroke survivors with mild and moderate impairment. Frontiers in Human Neuroscience, 9, 6.

83.

Roh

, Rymer

W.Z.

, Perreault

E.J.

, Yoo

S.B.

, & Beer

R.F.

(2013). Alterations in upper limb muscle synergy structure in chronic stroke survivors. Journal of Neurophysiology, 109(3), 768–781.

84.

Rose

D.K.

, & Winstein

C.J.

(2004). Bimanual training after stroke: Are two hands better than one? Topics in Stroke Rehabilitation, 11(4), 20–30.

85.

Rose

D.K.

, & Winstein

C.J.

(2005). The co-ordination of bimanual rapid aiming movements following stroke. Clinical Rehabilitation, 19(4), 452–462.

86.

Rose

D.K.

, & Winstein

C.J.

(2013). Temporal coupling is more robust than spatial coupling: An investigation of interlimb coordination after stroke. Journal of Motor Behavior, 45(4), 313–324.

87.

Sainburg

, Good

, & Przybyla

(2013). Bilateral synergy: A framework for post-stroke rehabilitation. Journal of Neurology & Translational Neuroscience, 1(3), pii: 1025.

88.

Sathian

, Buxbaum

L.J.

, Cohen

L.G.

, Krakauer

J.W.

, Lang

C.E.

, Corbetta

, & Fitzpatrick

S.M.

(2011). Neurological principles and rehabilitation of action disorders: Common clinical deficits. Neurorehabilitation and Neural Repair, 25(5 Suppl), 21S–32S.

89.

Scholz

J.P.

(1990). Dynamic pattern theory–some implications for therapeutics. Physical Therapy, 70(12), 827–843.

90.

Shah-Basak

P.P.

, Norise

, Garcia

, Torres

, Faseyitan

, & Hamilton

R.H.

(2015). Individualized treatment with transcranial direct current stimulation in patients with chronic non-fluent aphasia due to stroke. Frontiers in Human Neuroscience, 9, 201.

91.

Shea

C.H.

, Buchanan

J.J.

, & Kennedy

D.M.

(2016). Perception and action influences on discrete and reciprocal bimanual coordination. Psychonomic Bulletin & Review, 23(2), 361–386.

92.

Shirota

, Jansa

, Diaz

, Balasubramanian

, Mazzoleni

, Borghese

N.A.

, & Melendez-Calderon

(2016). On the assessment of coordination between upper extremities: Towards a common language between rehabilitation engineers, clinicians and neuroscientists. Journal of Neuroengineering and Rehabilitation, 13(1), 80.

93.

Sleimen-Malkoun

, Temprado

J.-J.

, Thefenne

, & Berton

(2011). Bimanual training in stroke: How do coupling and symmetry-breaking matter? BMC Neurology, 11(1), 11.

94.

Stinear

C.M.

, Barber

P.A.

, Coxon

J.P.

, Fleming

M.K.

, & Byblow

W.D.

(2008). Priming the motor system enhances the effects of upper limb therapy in chronic stroke. Brain, 131(5), 1381–1390.

95.

Stinear

C.M.

, Petoe

M.A.

, Anwar

, Barber

P.A.

, & Byblow

W.D.

(2013). Bilateral priming accelerates recovery of upper limb function after stroke: A randomized controlled trial. Stroke, 45(1), 205–210.

96.

Stoykov

M.E.

, Lewis

G.N.

, & Corcos

D.M.

(2009). Comparison of bilateral and unilateral training for upper extremity hemiparesis in stroke. Neurorehabilitation and Neural Repair, 23(9), 945–953.

97.

Stoykov

M.E.

, & Madhavan

(2015). Motor priming in neurorehabilitation. Journal of Neurologic Physical Therapy: JNPT, 39(1), 33–42.

98.

Stoykov

M.E.

, & Stinear

J.W.

(2010). Active-passive bilateral therapy as a priming mechanism for individuals in the subacute phase of post-stroke recovery: A feasibility study. American Journal of Physical Medicine & Rehabilitation, 89(11), 873–878.

99.

Subramanian

S.K.

, Lourenço

C.B.

, Chilingaryan

, Sveistrup

, & Levin

M.F.

(2013). Arm motor recovery using a virtual reality intervention in chronic stroke: Randomized control trial. Neurorehabilitation and Neural Repair, 27(1), 13–23.

100.

Swinnen

S.P.

(2002). Intermanual coordination: From behavioural principles to neural-network interactions. Nature Reviews. Neuroscience, 3(5), 348–359.

101.

Swinnen

S.P.

, & Wenderoth

(2004). Two hands, one brain: Cognitive neuroscience of bimanual skill. Trends in Cognitive Sciences, 8(1), 18–25.

102.

Takahashi

C.D.

, & Reinkensmeyer

D.J.

(2003). Hemiparetic stroke impairs anticipatory control of arm movement. Experimental Brain Research, 149(2), 131–140.

103.

Takeuchi

, Tada

, Matsuo

, & Ikoma

(2012). Low-frequency repetitive TMS plus anodal transcranial DCS prevents transient decline in bimanual movement induced by contralesional inhibitory rTMS after stroke. Neurorehabilitation and Neural Repair, 26(8), 988–998.

104.

Tazoe

, Sasada

, Sakamoto

, & Komiyama

(2013). Modulation of interhemispheric interactions across symmetric and asymmetric bimanual force regulations. European Journal of Neuroscience, 37(1), 96–104.

105.

Tretriluxana

, Kantak

, Tretriluxana

, Wu

A.D.

, & Fisher

B.E.

(2013). Low frequency repetitive transcranial magnetic stimulation to the non-lesioned hemisphere improves paretic arm reach-to-grasp performance after chronic stroke. Disability and Rehabilitation. Assistive Technology, 8(2), 121–124.

106.

van Delden

A.L.E.Q.

, Peper

C.L.E.

, Harlaar

, Daffertshofer

, Zijp

N.I.

, Nienhuys

, & … Beek

P.J.

(2009). Comparing unilateral and bilateral upper limb training: The ULTRA-stroke program design. BMC Neurology, 9, 57.

107.

van Delden

A.E.Q.

, Beek

P.J.

, Roerdink

, Kwakkel

, & Peper

C.E.

(2014). Unilateral and bilateral upper-limb training interventions after stroke have similar effects on bimanual coupling strength. Neurorehabilitation and Neural Repair, 29(3), 255–267.

108.

Vinjamuri

, Sun

, Crammond

, Sclabassi

, & Mao

Z.-H.

(2008). Inherent bimanual postural synergies in hands. Conference Proceedings:... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, 2008, 5093–5096.

109.

Waddell

K.J.

, Strube

M.J.

, Bailey

R.R.

, Klaesner

J.W.

, Birkenmeier

R.L.

, Dromerick

A.W.

, & Lang

C.E.

(2017). Does task-specific training improve upper limb performance in dailylife poststroke? Neurorehabilitation and NeuralRepair, 31(3), 290–300.

110.

Wade

D.T.

, Langton-Hewer

, Wood

V.A.

, Skilbeck

C.E.

, & Ismail

H.M.

(1983). The hemiplegic arm after stroke: Measurement and recovery. Journal of Neurology, Neurosurgery, and Psychiatry, 46(6), 521–524.

111.

Wagner

J.M.

, Lang

C.E.

, Sahrmann, S. a

Hu, Q.

, Bastian

A.J.

, Edwards

D.F.

, & Dromerick

A.W.

(2006). Relationships between sensorimotor impairments and reaching deficits in acute hemiparesis. Neurorehabilitation and Neural Repair, 20(3), 406–416.

112.

Whitall

McCombe

Waller S.

Silver

K.H.

, & Macko

R.F.

(2000a). Repetitive bilateral arm training with rhythmic auditory cueing improves motor function in chronic hemiparetic stroke. Stroke; a Journal of Cerebral Circulation, 31(10), 2390–2395.

113.

Whitall

, Waller

S.M.

, Sorkin

J.D.

, Forrester

L.W.

, Macko

R.F.

, Hanley

D.F.

, & … Luft

(2011). Bilateral and unilateral arm training improve motor function through differing neuroplastic mechanisms: A single-blinded randomized controlled trial. Neurorehabilitation and Neural Repair, 25(2), 118–129.

114.

Winstein

C.J.

, Stein

, Arena

, Bates

, Cherney

L.R.

, Cramer

S.C.

, & … American Heart Association Stroke Council

Council on Cardiovascular and Stroke Nursing, Council on Clinical Cardiology, and Council on Quality of Care and Outcomes Research.

(2016). Guidelines for Adult Stroke Rehabilitation and Recovery: A Guideline for Healthcare Professionals From the American Heart Association/American Stroke Association. Stroke; a Journal of Cerebral Circulation, 47(6), e98–e169.

115.

Wolf

, Scheiderer

, Napolitan

, Belden

, Shaub

, & Whitford

(2014). Efficacy and task structure of bimanual training post stroke: A systematic review. Topics in Stroke Rehabilitation, 21(3), 181–196.

116.

Wolf

S.L.

(2007). Revisiting constraint-induced movement therapy: Are we too smitten with the mitten? Is all nonuse “learned”? and other quandaries. Physical Therapy, 87(9), 1212–1223.

117.

Wolf

S.L.

, Thompson

P.A.

, Winstein

C.J.

, Miller

J.P.

, Blanton

S.R.

, Nichols-Larsen

D.S.

, & … Sawaki

(2010). The EXCITE stroke trial: Comparing early and delayed constraint-induced movement therapy. Stroke, 41(10), 2309–2315.

118.

Wolf

S.L.

, Winstein

C.J.

, Miller

J.P.

, Taub

, Uswatte

, Morris

, & … Y

(2006). Effect of constraint-induced movement therapy on upper extremity function 3 to 9 months after stroke. JAMA, 296(17), 2095.

119.

C.-Y.

, Chou

S.-H.

, Chen

C.-L.

, Kuo

M.-Y.

, Lu

T.-W.

, & Fu

Y.-C.

(2009). Kinematic analysis of a functional and sequential bimanual task in patients with left hemiparesis: Intra-limb and interlimb coordination. Disability and Rehabilitation, 31(12), 958–966.

120.

, Lin

, Chen

, & Hong

(2007). Effects of modified constraint-induced movement therapy on movement kinematics and daily function in patients with stroke: A kinematic study of motor control mechanisms. Neurorehabilitation and Neural Repair, 21(5), 460–466.

121.

Yokoi

, Hirashima

, & Nozaki

(2011). Gain field encoding of the kinematics of both arms in the internal model enables flexible bimanual action. Journal of Neuroscience, 31(47), 17058–17068.