Robot-assisted voluntary initiation reduces control-related difficulties of initiating joint movement: A phenomenal questionnaire study on shaping and compensation of forward gait

Patrick Grüneberg; Hideki Kadone; Naomi Kuramoto; Tomoyuki Ueno; Yasushi Hada; Masashi Yamazaki; Yoshiyuki Sankai; Kenji Suzuki

doi:10.1371/journal.pone.0194214

Abstract

Humans employ various control strategies to initiate and maintain bodily movement. In case that the normal gait function is impaired, exoskeleton robots provide motor assistance during therapy. While the robotic control system builds on kinematic gait functions, the patient’s voluntary efforts to initiate motion also contribute to the effectiveness of the therapy process. However, it is currently not well understood how voluntary initiation as a subjective capacity affects the physiological level of motor control. In order to understand the functional nexus between voluntary initiation and motor control, we interviewed patients undergoing robotic gait rehabilitation with the HAL exoskeleton robot about their experience and command of voluntarily initiating forward gait while using the HAL system. Their reports provide phenomenal evidence for voluntary initiation as a distinct cognitive act that comes as phenomenal performance. Furthermore, phenomenal evidence about the functional relation of intention and initiation correlates with FIM-M gait scores. Based on the assumption that HAL reduces control-related difficulties of voluntarily initiating joint movement, we identified two cognitive control strategies, shaping and compensation of gait, that imply a heterarchic organization of the human system of action control.

Citation: Grüneberg P, Kadone H, Kuramoto N, Ueno T, Hada Y, Yamazaki M, et al. (2018) Robot-assisted voluntary initiation reduces control-related difficulties of initiating joint movement: A phenomenal questionnaire study on shaping and compensation of forward gait. PLoS ONE 13(3): e0194214. https://doi.org/10.1371/journal.pone.0194214

Editor: Yih-Kuen Jan, University of Illinois at Urbana-Champaign, UNITED STATES

Received: October 24, 2017; Accepted: February 27, 2018; Published: March 12, 2018

Copyright: © 2018 Grüneberg et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: Protocol via protocols.io: dx.doi.org/10.17504/protocols.io.mqnc5ve. Questionnaire via figshare: dx.doi.org/10.6084/m9.figshare.5782812. Subject data and results via figshare: dx.doi.org/10.6084/m9.figshare.5782881.

Funding: The authors received no specific funding for this work.

Competing interests: The authors have declared that no competing interests exist.

Introduction

The purpose of robotic systems for therapy of upper [1] and lower limbs [2] is to support patients in executing motions that they cannot anymore execute by themselves. Recent advances in this new field of robot-assisted therapy suggest that the active participation of the patient in the therapeutic process has positive effects on motor recovery [3]. I.e. the patient’s voluntary involvement while using a robotic device supports neuroplasticity for therapeutic purposes and motor learning. However, voluntary resources are not considered in the standard electromechanical approach to robot-assisted therapy. Most exoskeleton robots (as well as traditional physiotherapy) act on the (distal) physical level in order to influence the neural system. This approach leaves the patient passive: the impaired limbs of a patient are being moved.

In order to fully exploit the active participation of the patient, an alternative control strategy builds on biosignals [4] within the proprioceptive loop of movement initiation, execution and feedback to the brain. Even in the case of severely impaired patients, efferent EMG signals of neural muscle activity often remain in the muscles, but do not anymore result in sufficient joint movement. The exoskeleton robot HAL (hybrid assistive limbs; Fig 1) detects and interprets these signals [5] as the patient’s intention to move his body voluntarily [6]. Accordingly, the patient activates the robot by means of voluntary initiation [VI] of forward gait [7], instead of being moved passively by predetermined kinematic patterns. Preliminary studies of HAL-assisted therapy provide provisional clinical data regarding beneficial effects on gait therapy [8] ranging from stroke patients [9][10][11][12][13] to patients with spinal cord infarction, musculoskeletal and other diseases [14][15][16] and one patient with ossification of the posterior longitudinal ligament (OPLL) [17].

Download:

Fig 1. HAL patient during therapy.

https://doi.org/10.1371/journal.pone.0194214.g001

Currently, HAL works with a broad variety of impairments. Regarding the interaction between patient and robot, any EMG-signals in the leg muscles–regardless the respective diagnosis–are interpreted in a standardized manner as the patient’s intention to move. Despite the implementation of HAL, it is still unknown how VI actually operates on motor programs. In the face of increasing evidence that volitional capabilities play an efficacious role in movement control, the importance of investigating subjective execution of volitional acts has been stressed recently [18][19]. Whereas such volitional capabilities are usually ascribed to a human subject of movement [20], current research in movement and its behavioral underpinnings is mostly concerned with kinematic aspects [21]. In turn, philosophical accounts of volition [22] do not focus on the physical implementation of volitional acts. Phenomenological psychology of the early 20^th century investigated VI as a distinct phenomenal act [23][24][25][26], but vanished in the advent of behaviorism. Also current phenomenological and cognitive (neuro)science research on bodily movement focuses merely on different types of sense of agency and ownership [27][28][29]. Yet, any sense of agency does not cover the genuine act of initiating a movement that is left to subpersonal processes in the brain and not to the conscious subject. In sum, currently hardly any account or data simultaneously considers the subjective dimension of VI as an efficacious capacity and its physical implementation in joint movement. The functional relation between volitional acts and bodily movement is rather omitted.

In order to understand how VI operates on motor programs, it is necessary to investigate the subjective (first-person) perspective of the human user during VI of bodily movement. Opposed to healthy subjects who usually do not have a particular consciousness of movement initiation [30], HAL patients have experiences of “being able to move” (before their impairment), “not being able to move” (due to their impairment) and “again (partially) being able to move” (in the course and after therapy). Based on these diverging experiences, they are qualified to identify VI as a particular cognitive act and specify its role in the system of action control. Whereas introspective accounts were disregarded in the sense that introspectively described states or acts do not exhibit any functional (causal) relevance [31], the HAL setting renders these reports relevant in that these describe efficacious phenomenal behavior [7]. Based on these reports and the functional role of VI in the proprioceptive loop of initiation, execution and feedback of bodily movement, we aim to test a set of two hypotheses:

H₁: VI comes as a phenomenal performance.
H₂: VI is a cognitively distinct act of action control.

For this purpose, we interviewed HAL patients about their first-person execution of VI while controlling the HAL robot during therapy sessions. We then combined the phenomenal (subjective) evidence for VI with objectively measured gait scores. Hereby, we introduced a novel method for modelling subjective acts (VI) in relation to their physical implementation in joint movement showing that the subjective (voluntary) conditions of the patients correlates with their physiological motor functions. With this method, it was possible to confirm VI as a conscious and cognitively distinct act as well as to specify two cognitive control strategies that patients employ in order to initiate forward gait. These strategies imply two respective modes of robot-assisted VI, revealing that HAL reduces control-related difficulties of voluntarily initiating joint movement.

Results

We collected data by means of interviewing 20 HAL patients at University of Tsukuba Hospital. After the patients completed their HAL therapy program, we used a multiple-choice questionnaire with an additional option for comments to every question in order to investigate the first-person (phenomenal) experience while using the HAL robot. The questionnaire [32] consists of totally 25 questions, grouped in four items. In a second step, the results of the questionnaire were correlated to gait scores pre and post therapy (see Materials and methods for more details). The subjects’ age range from one young patient (12 years) to one 79 year-old patient with average age of 59.1 years (± 16.58; Fig 2). The conditions of the subjects range from stroke induced diagnoses to spinal cord infarction and neurological diseases (Table 1). All diagnoses impaired the gait function according to the International Classification of Functioning, Disability and Health (ICF b760 Control of voluntary movement functions; b770 Gait pattern functions; d450 Walking). Subjects joined the therapy program during acute and chronic stages (time span between diagnosis and start of HAL sessions lies between less than one month and more than one year; Fig 3A) whereby this condition did not affect the results. The rehabilitation program included average 8.45 (±1.87) HAL sessions per patient (Fig 3B) and average 3.8 (±1.79) weeks of HAL sessions per patient (Fig 3C). During the interview, subjects did not always answer to all questions. In some cases, qualitative answers provided additional information. Due to the explorative nature of this study, not all items provided results relevant for the hypotheses (for an overview of all results see [33]).

Download:

Fig 2. Age and gender of subjects.

12 male, 8 female subjects; n = 20, mean age 59.1, SD ± 16.58 years.

https://doi.org/10.1371/journal.pone.0194214.g002

Download:

Fig 3. Histogram of participation in therapy.

(A) Time span between diagnosis and start of HAL sessions (n = 20). (B) Total number of HAL sessions (n = 20, mean 8.45, SD ±1.87). (C) Number of weeks of HAL sessions (n = 20, mean 3.8, SD ±1.79).

https://doi.org/10.1371/journal.pone.0194214.g003

Download:

Table 1. Diagnoses of subjects.

https://doi.org/10.1371/journal.pone.0194214.t001

Voluntary initiation as phenomenal performance

All subjects confirm that they intentionally participated in the HAL therapy (item 1.1: p < .00001; Table 2), that they were willing to walk forward (item 1.2: p < .00001; Table 2) and that they initiated forward walking by themselves (item 1.3: p < .00001; Table 2). Thus, from a phenomenal first-person perspective, subjects are fully aware of VI. Regarding the phenomenal quality, VI involves motor imagery [34] (item 2.1: p = .00002; Table 3). More specifically, subjects report about dynamic mental imagery, i.e. they have an internal image of the sequence of their walking motion while executing the movement [35] (item 2.2: p = .0033; Table 3). Regarding the phenomenal type, 10 subjects characterized VI as a phenomenal performance (item 2.4: p = .00008; Table 3); 7 subjects rely on thought (p = .00967) that occurs in two cases simultaneously with the performance. E.g., subject 11 explains that she thinks and judges whether her motion is appropriate while initiating movement. Taken together, items 1.1–3, 2.1/2/4 suggest phenomenal evidence for H₁. Other phenomenal types of action-related consciousness did not prove to be significant, such as one case of decision-making (p = .16894) that co-occurred with thought and one case of memory (p = .16894).

Download:

Table 2. Results of item group 1.

https://doi.org/10.1371/journal.pone.0194214.t002

Download:

Table 3. Results of item group 2.

https://doi.org/10.1371/journal.pone.0194214.t003

Voluntary initiation as a distinct act in the system of action control

Subjects distinguished VI from preceding (intention and urge) and subsequent (joint movement, emotion and motivation) functions of action control. Regarding preceding functions, the identification of VI as a distinct act occurs regardless whether the ability to initiate movement voluntarily depends on the preceding intention [36][37] to join rehabilitation (item 3.1: p = .00739; Table 4) or whether a subject can initiate movement independently of her intention to join therapy (p = .00739). The same applies for the relation between a preceding urge [38] and VI. Subjective evidence strongly suggests an inner urge to walk forward after subjects have been prepared with HAL and are ready to move (item 3.4: p = .00018; Table 4). On the other hand, VI does not necessarily follow from the urge because subjects state that they can do nothing about the urge and just have to walk forward (item 3.5: p = .09442; Table 4) as well as that they initiate forward walking independently of the urge (p = .09442). Thus, VI maintains a distinct functional role during action control because the preceding intention and the occurrence of an urge do not necessarily imply the execution of joint movement.

Download:

Table 4. Results of item group 3.

https://doi.org/10.1371/journal.pone.0194214.t004

Regarding subsequent functions, subjects distinguish between VI and the implementation of bodily movement. For the execution of VI, it seems to be irrelevant whether the body moves as desired (item 3.2: p = .12013; Table 4) or not as desired (p = .12013). Furthermore, subjective exercise of VI can occur independently of an actual movement (item 3.3: p = .00011; Table 4). Finally, the data suggests that VI affects emotion positively (item 3.6: p < .00001; Table 4) and increases motivation for therapy [39] (item 3.7: p < .00001; Table 4). In turn, the data does not suggest that there is no effect on emotion (item 3.6: p = .06695) and motivation (item 3.7: p = .1339). Subjects confirm a positive impact of successful movement on motivation (item 3.8: p < .00001; Table 4) and the corresponding increase of motivation (item 3.9: p = .00002; Table 4). In sum, subjective evidence indicates VI as a cognitively distinct act in the system of action control (H₂).

Two groups of robot-assisted voluntary initiation

Based on H₁ and H₂, a matching of phenomenal evidence for VI with the gait scores of the subjects derived a correlation between the relation of VI to intention (item 3.1) and the FIM-M pre-therapy gait score. Results imply two groups (X² = 10.322, df = 1, p = .00131; Table 5): Members of group A show a pre-therapy FIM-M > 24 and explain that they can voluntarily initiate forward walking independently of the preceding intention to join therapy (item 3.1: p = .00739; Table 4). Members of group B show a pre-therapy FIM-M ≤ 24 and explain that their ability to initiate forward walking depends on their intention to join therapy (item 3.1: p = .00739). No member of group B has a higher score of FIM-M pre therapy than any member of group A. According to our data, FIM-M ≤ 24 seems to imply a cognitive border below which VI alone is insufficient for initiating voluntary movement (however, the exact value of this border may vary in other data sets). The correlation between VI and FIM-M holds for the condition of the subjects pre-therapy (Fig 4A) and regardless of FIM-C pre-therapy (Fig 4B) or age (Fig 4C).

Download:

Fig 4. Distribution of FIM-M over two groups of robot-assisted initiation.

(A) Distribution of FIM-M scores pre- and post-therapy regarding group A (FIM-M > 24) and group B (FIM-M ≤ 24) (Group A pre: mean 58.533, SD 24.053, post: mean 73.067, SD 16.867; Group B pre: mean 18, SD 5.568; post: mean 54.333, SD 21.385; * n = 20, X² = 10.322, df = 1, p = .00131). (B) Distribution of FIM-M (0–91) and FIM-C (0–35) scores over group A and B. (C) Distribution of FIM-M (0–91) and age over group A and B.

https://doi.org/10.1371/journal.pone.0194214.g004

Download:

Table 5. Correlation of FIM-M pre therapy with item 3.1 (relation of initiation to intention).

https://doi.org/10.1371/journal.pone.0194214.t005

Discussion

Subjective evidence for voluntary initiation of joint movement

Results building on introspective reports, particularly in the case of subjects with cognitive impairments, raise questions about the reliability of the responses. With respect to the contents of the responses, the results of this study conform with the acknowledged contribution of motor imagery to the therapeutic process [35][40]. With respect to the method, first, medical doctors recommended all subjects and qualified them as being capable to participate. Second, in most cases FIM-C was rather high. Even in the cases of lower FIM-C, there were no cases of impossible communication or misunderstanding during the interviews. In sum, there seem to be no substantial limitations to the reliability of the responses although a higher number of subjects with a low FIM-M is necessary in order to strengthen the support for the hypotheses.

A further possible bias of the results concerns the comparability of subjects with different disorders.

However, from a clinical viewpoint, the question is (independently of the specific disorder and the disease process) whether a patient is willing to initiate movement or not. Accordingly, the questionnaire focuses on the particular cognitive issue of VI of joint movement and does not aim at evaluating physical performance (FIM-M) or other cognitive issues (FIM-C). Furthermore, diagnoses and disease stage did not affect the responses. Thus, subjects are treated equally only with respect to VI so that comparability within a heterogeneous population could be reasonably assumed.

Efficacious action consciousness

Results regarding the phenomenal quality and type suggest that VI does not fit in the available phenomenological classification of voluntary acts [41][27][42][43] and implies a genuine type of voluntary action. This subjective evidence confirms the theoretical model of VI [7], according to which VI comes as efficacious action consciousness that bears phenomenal content of the immediate execution (performance) of an action (H₁). Regarding its individuation (H₂), VI as the “conscious initiation (the implementation of a decision or intention to act in actual behavior)” can be distinguished from the intention as “practical decision-making (deliberation and termination of deliberation in favor of one possible action)” [44] and other functions related to action control.

In contrast, neuropsychological approaches to action control do not ascribe any efficacious function to volition. According to the WWW model [45], intentional action builds on decisions about which action to execute (what component), when to execute an action (when component) and whether to execute an action (whether component). This classification does not clearly specify VI in the functional terms as suggested by the HAL scenario because the what component is part of action planning. Therefore, it comes as what we specified as intention and does not bear any immediate efficacy on joint movement. The when component is an essential aspect of VI, but it is not clear in the WWW model whether timing includes the actual execution of movement. The whether component is too narrowly defined in terms of intentional inhibition of action. Neither can intentional accounts capture VI as a phenomenal and efficacious capacity. Based on attempts to reduce initiation to intention, different classes of intentions are suggested [36][37]. Distal intentions are concerned with the nonimmediate future, proximal intentions are supposed to exhibit their causal role on behavior through intention acquisition, and motor intentions directly work on the physical motor system. However, due to the representational stance, intentions do not bear the functional capacity to operate on motor programs [7][22].

Heterarchic organization of action control

The phenomenal and causal characteristics of VI give rise to questions about the functional organization of the system of action control. In particular, the results of item 3.1 are relevant because they correlate with the objective condition of the subjects and thereby describe a relation between VI and the motor level. According to the phenomenal report and the FIM-M pre-therapy value, two cognitive control strategies and two respective modes of robot-assisted initiation can be distinguished based on the three functions of (1) intention, (2) VI and (3) joint movement (Fig 5):

Shaping: Subjects of group A execute VI to the extent that they are able to initiate movement independently of any previous intention. This subjective condition correlates with a FIM-M > 24. Because these subjects control HAL by means of VI, HAL supports the subsequent implementation of joint movement, i.e. shapes their gait function.
Compensation: Subjects of group B cannot execute VI as an independent capacity. Instead, HAL enables these subjects to utilize additional intentional efforts in order to initiate joint movement and thereby compensates for insufficient VI.

Download:

Fig 5. Neural and voluntary pathways of shaping and compensation.

The red arrow in the neural path shows the functional location of gait impairment. The dotted line in the voluntary path of shaping (group A) shows that intention is present, but not necessary for initiating joint movement.

https://doi.org/10.1371/journal.pone.0194214.g005

The two modes suggest different configurations of action control in a heterarchic system. Opposed to common hierarchical views of action control where the governing function is predetermined, intention, initiation and joint movement do not stand in a linear, monocausal nexus from the former to the latter. In a heterarchic system, a configuration consists of the reciprocal activity of independent functions. Depending on the situation and needs of the agents, one of these functions governs the overall system [46][47] so that an agent is able to employ intention and initiation simultaneously (compensation) or independently (shaping).

In general, there is hardly any related work on cognitive control strategies of joint movement. The identification of voluntary gait patterns (opposed to automatic gait patterns) during the activation of gait synergies [48][49] suggests preliminary physical evidence for VI as the cognitive origin of these voluntary patterns. Additionally, phenomenological analysis of introspection by Ach can serve to clarify the employment of the two modes [23][50][51]. The “law of difficulty” states that voluntary effort increases as a positive function of the perceived difficulty of implementing an intended action [52] and thereby links the subjective perspective with the physical condition of the agent. Accordingly, HAL reduces control-related difficulties: subjects of both groups are enabled to execute VI despite their impairment whereby subjects of group B employ additional intentional efforts. Further investigation has to specify to what extent the impairment of members of group B originates in VI or the motor level respectively. The report of all members of group B about VI as a phenomenal performance (item 2.4) suggests the assumption that they actually execute VI. Hence, the impairment may be rather due to motor defects that disable VI.

Clinical implications

Knowledge about modes of efficacy of VI could provide more fine-grained criteria for the interpretation of EMG-signals through the robot. For this purpose, it would be necessary to identify variations in gait patterns and related EMG-signals according to the cognitive control strategies. This refinement could deepen the integration of robotic devices in the execution of volitional acts and allow for a more individualized robotic support of specific impairments.

Materials and methods

HAL therapy session

HAL (Hybrid Assistive Limb, Cyberdyne, Tsukuba, Japan) is a wearable type assistive exoskeleton robot for the lower limbs. It has segments to be attached to the waist, thigh, shank and foot. The single leg version of the robot was applied for stroke patients, and dual legs version for the others. The robot has electric motors to assist joint motion of hip and knee. The motors are actuated in one of the two modes: CVC (Cybernic Voluntary Control) and CAC (Cybernic Autonomous Control) mode. In CVC mode, the motors are driven according to neuro-muscular activation detected as electric signals measured using electrodes attached on the skin surface corresponding to the muscles for each of flexion and extension of hip and knee joint. By amplifying the detected electric activation as joint torque, CVC mode can support the lower limb motion in real-time according to the motion intention of the user. In CAC mode, the motors are controlled to reproduce a pre-defined trajectory according to foot-floor contact measured by floor reaction pressure sensor in the shoes. CAC mode is used for patients with complete loss of the electric activation on the lower limbs. CVC mode was applied to all of the patients in this study. Parameters for tuning the amplification gain was adjusted manually on site for each patient’s comfort and smoothness of gait.

One session of HAL therapy took 90 minutes, including clinical assessments before using HAL, attaching HAL, walking with HAL, detaching HAL and assessments after the walking training. Time duration of walking with HAL was 10–20 minutes in total in one session. A medical doctor, one or two physical therapists, and an assistant attended each session. For safety reasons, a walking device (All-in-One walking trainer, Ropox A/S, Naestved, Denmark) with a harness protected patients from falls during walking with HAL.

Experimental procedure

For each of the patients, an interview-based questionnaire was conducted just after finishing the last HAL session [53]. Before conducting the interview, the purpose, method, expected contribution, and privacy protection issues regarding the questionnaire were explained by a medical doctor before written informed consent was obtained. For some of the patients, typically stroke patients who participated in their acute phase, there was a concern that they might have not understood the explanation thoroughly. For these patients, their spouses gave written informed consent. The same experimenter did the interview for all of the patients. The experimenter first asked the patients whether they wanted to read the questionnaire sheet and fill in the answers by themselves. In this case, the experimenter stayed still beside the patients and, in case of any question, supported the patients’ understanding of the questionnaire. Otherwise, the experimenter verbally read out each of the sentences of the questionnaire and filled in the answers based on the patients’ verbal response. Ethical clearance for this research was reviewed and approved by the institutional review board of University of Tsukuba Hospital (Application No. H27-034).

Interview-based questionnaire

In order to learn about the subjective condition during VI of joint movement, it is necessary to ask subjects about their cognitive behavior in their first-person experience. The relevant method here is phenomenological introspection. While there are still substantial doubts about the reliability of introspection [31], it is the only available method for this purpose which has been developed extensively in phenomenological psychology [23][50][51][26]. In contrast to qualitative interview approaches which build on open-ended questions [18], our questionnaire builds on multiple-choice questions (with the opportunity to give remarks at every question) for the sake of quantification [32]. During the interview, item 2.4, concerning the overall phenomenal quality of VI, appears as a summarizing question Q9 after Q8.2 before passing over to item group 4 on control issues.

The questionnaire contains the following items:

Subjective experience of voluntary initiation
1. 1.1. Intention (Q1)
2. 1.2. Volition (Q2)
3. 1.3. Initiation (Q3)
Phenomenal quality of voluntary initiation
1. 2.1. Imagery (Q4.1)
2. 2.2. Type of imagery (Q4.2)
3. 2.3. Non-pictorial awareness of body parts (Q4.3)
4. 2.4. Experiential quality (Q9)
Functional embedding in action control
1. Intention
2. 3.1. Relation of initiation to intention (Q5.1)
3. 3.2. Body movement during rehabilitation (Q5.2)
4. 3.3. Relation of intended movement and initiation (Q5.3)
  Urge
5. 3.4. Occurrence of urge (Q6.1)
6. 3.5. Relation of urge to initiation (Q6.2)
  Motivational system
7. 3.6. Effects on emotion (Q7.1)
8. 3.7. Effects on motivation (Q7.2)
9. 3.8. Rehabilitation and motivation (Q8.1)
10. 3.9. Changes of motivation (Q8.2)
Control issues
1. 4.1. Mental and physical effort (Q10.1, 10.2, 10.3, 10.4, 10.5, 10.6)
2. 4.2. Futile efforts to move (Q11)
3. 4.3. Walking speed (Q12)
4. 4.4. Walking direction (Q13)

Subjects and statistical data

Twenty subjects participated in the study. The participants were patients at the University of Tsukuba Hospital. All of them had locomotor dysfunction due to damages in the central nervous system. The patients showed different diagnoses for the principal cause of motor impairment. The remaining locomotor function depended on the cause, severity of damage and time from diagnosis. Typically, stroke patients had hemiparesis, and myelopathy patients had paraparesis. Statistical data includes the following:

Gender: Male / Female
Age
Diagnosis
1. ○. Cerebral disorder
2. ○. Spinal disorder
Time span between diagnosis and start of HAL sessions
Total number of HAL sessions
Number of weeks of HAL sessions

Availability of stroke patients who were capable of participating in the interview was highly limited.

Clinical assessment

For each patient, clinical assessment was conducted before starting the first HAL session and after finishing the last HAL session. The clinical assessment included evaluation according to FIM (Functional Independence Measure) [54] separately for motor related scores (FIM-M) and cognition related scores (FIM-C), mRS (modified Rankin Scale) [55] and BI (Barthel Index) [56], in addition to a standard 10m or 6m walking test.

References

1. Maciejasz P, Eschweiler J, Gerlach-Hahn K, Jansen-Troy A, Leonhardt S. A survey on robotic devices for upper limb rehabilitation. J NeuroEngineering Rehabil. 2014 Jan 9;11(1):3.
- View Article
- Google Scholar
2. Díaz I, Gil JJ, Sánchez E. Lower-Limb Robotic Rehabilitation: Literature Review and Challenges. J Robot. 2011 Nov 17;2011:e759764.
- View Article
- Google Scholar
3. Hogan N, Krebs HI, Rohrer B, Palazzolo JJ, Dipietro L, Fasoli SE, et al. Motions or muscles? Some behavioral factors underlying robotic assistance of motor recovery. J Rehabil Res Dev. 2006;43(5):605. pmid:17123202
- View Article
- PubMed/NCBI
- Google Scholar
4. Belda-Lois J-M, Horno SM, Bermejo-Bosch I, Moreno JC, Pons JL, Farina D, et al. Rehabilitation of gait after stroke: a review towards a top-down approach. J NeuroEngineering Rehabil. 2011 Dec 13;8(1):66.
- View Article
- Google Scholar
5. de Almeida Ribeiro PR, Brasil FL, Witkowski M, Shiman F, Cipriani C, Vitiello N, et al. Controlling Assistive Machines in Paralysis Using Brain Waves and Other Biosignals. Adv Hum-Comp Int. 2013 Jan;2013:3:3–3:3.
- View Article
- Google Scholar
6. Suzuki K, Mito G, Kawamoto H, Hasegawa Y, Sankai Y. Intention-based walking support for paraplegia patients with Robot Suit HAL. In: Miripour B, editor. Climbing and Walking Robots. Rijeka, Shanghai: InTech; 2007. p. 1441–69.
7. Grüneberg P, Kadone H, Suzuki K. Voluntary initiation of movement: multifunctional integration of subjective agency. Front Psychol. 2015;6:688. pmid:26052308
- View Article
- PubMed/NCBI
- Google Scholar
8. Wall A, Borg J, Palmcrantz S. Clinical application of the Hybrid Assistive Limb (HAL) for gait training—a systematic review. Front Syst Neurosci. 2015 Mar 25;9.
- View Article
- Google Scholar
9. Maeshima S, Osawa A, Nishio D, Hirano Y, Takeda K, Kigawa H, et al. Efficacy of a hybrid assistive limb in post-stroke hemiplegic patients: a preliminary report. BMC Neurol. 2011 Dec;11(1).
- View Article
- Google Scholar
10. Ueba T, Hamada O, Ogata T, Inoue T, Shiota E, Sankai Y. Feasibility and Safety of Acute Phase Rehabilitation After Stroke Using the Hybrid Assistive Limb Robot Suit. Neurol Med Chir (Tokyo). 2013;53(5):287–90.
- View Article
- Google Scholar
11. Kawamoto H, Kamibayashi K, Nakata Y, Yamawaki K, Ariyasu R, Sankai Y, et al. Pilot study of locomotion improvement using hybrid assistive limb in chronic stroke patients. BMC Neurol. 2013 Oct 7;13(1):141.
- View Article
- Google Scholar
12. Nilsson A, Vreede KS, Häglund V, Kawamoto H, Sankai Y, Borg J. Gait training early after stroke with a new exoskeleton–the hybrid assistive limb: a study of safety and feasibility. J Neuroengineering Rehabil. 2014;11(1):1.
- View Article
- Google Scholar
13. Watanabe H, Tanaka N, Inuta T, Saitou H, Yanagi H. Locomotion Improvement Using a Hybrid Assistive Limb in Recovery Phase Stroke Patients: A Randomized Controlled Pilot Study. Arch Phys Med Rehabil. 2014 Nov;95(11):2006–12. pmid:25010538
- View Article
- PubMed/NCBI
- Google Scholar
14. Kubota S, Nakata Y, Eguchi K, Kawamoto H, Kamibayashi K, Sakane M, et al. Feasibility of Rehabilitation Training With a Newly Developed Wearable Robot for Patients With Limited Mobility. Arch Phys Med Rehabil. 2013 Jun;94(6):1080–7. pmid:23313372
- View Article
- PubMed/NCBI
- Google Scholar
15. Aach M, Cruciger O, Sczesny-Kaiser M, Höffken O, Meindl RC, Tegenthoff M, et al. Voluntary driven exoskeleton as a new tool for rehabilitation in chronic spinal cord injury: a pilot study. Spine J. 2014 Dec;14(12):2847–53. pmid:24704677
- View Article
- PubMed/NCBI
- Google Scholar
16. Sczesny-Kaiser M, Höffken O, Aach M, Cruciger O, Grasmücke D, Meindl R, et al. HAL® exoskeleton training improves walking parameters and normalizes cortical excitability in primary somatosensory cortex in spinal cord injury patients. J NeuroEngineering Rehabil. 2015 Dec;12(1).
- View Article
- Google Scholar
17. Sakakima H, Ijiri K, Matsuda F, Tominaga H, Biwa T, Yone K, et al. A Newly Developed Robot Suit Hybrid Assistive Limb Facilitated Walking Rehabilitation after Spinal Surgery for Thoracic Ossification of the Posterior Longitudinal Ligament: A Case Report. Case Rep Orthop. 2013 Nov 28;2013:e621405.
- View Article
- Google Scholar
18. Nahmias E, Morris S, Nadelhoffer T. The Phenomenology of Free Will. J Conscious Stud. 2004;11(7–8):162–79.
- View Article
- Google Scholar
19. Pacherie E. Can Conscious Agency Be Saved? Topoi. 2014 Apr 1;33(1):33–45.
- View Article
- Google Scholar
20. Haggard P, Lau H. What is volition? Exp Brain Res. 2013 Sep 1;229(3):285–7. pmid:23727831
- View Article
- PubMed/NCBI
- Google Scholar
21. Götz-Neumann K. Gehen verstehen: Ganganalyse in der Physiotherapie. Stuttgart: Thieme; 2011.
22. Zhu J. The Conative Mind: Volition and Action. Saarbrücken: LAP LAMBERT Academic Publishing; 2012. 140 p.
23. Ach N. Über die Willenstätigkeit und das Denken. Eine experimentelle Untersuchung mit einem Anhange: Über das Hippsche Chronoskop. Göttingen: Vandenhoeck & Ruprecht; 1905.
24. Pfänder A. Phänomenologie des Wollens. Eine psychologische Analyse. Leipzig: J. A. Barth; 1900.
25. Lindworsky J. Der Wille: seine Erscheinung und seine Beherrschung nach den Ergebnissen der experimentellen Forschung. Leipzig: J.A. Barth; 1921.
26. Michotte A, Prum E. Le Choix Volontaire et ses antededents immediats. Arch Psychol. 1910;38–39:113–296.
- View Article
- Google Scholar
27. Synofzik M, Vosgerau G. Beyond the comparator model. Conscious Cogn. 2012 Mar;21(1):1–3. pmid:22301453
- View Article
- PubMed/NCBI
- Google Scholar
28. Chambon V, Sidarus N, Haggard P. From action intentions to action effects: how does the sense of agency come about? Front Hum Neurosci. 2014;8:320. pmid:24860486
- View Article
- PubMed/NCBI
- Google Scholar
29. Kühn S, Brass M, Haggard P. Feeling in control: Neural correlates of experience of agency. Cortex. 2013 Jul;49(7):1935–42. pmid:23067726
- View Article
- PubMed/NCBI
- Google Scholar
30. Metzinger T. Conscious Volition and Mental Representation: Toward a More Fine-Grained Analysis. In: Sebanz N, Prinz W, editors. Disorders of Volition. Cambridge, Mass.: MIT Press; 2006.
31. Horgan T, Timmons M. Introspection and the Phenomenology of Free Will: Problems and Prospects. J Conscious Stud. 2011;18(1):180–205.
- View Article
- Google Scholar
32. Grüneberg P, Kadone H, Kuramoto N, Suzuki K, Sankai Y. Voluntary initiation questionnaire JP-ENG. Figshare; 2018. Available from: dx.doi.org/10.6084/m9.figshare.5782812
33. Grüneberg P, Kadone H, Kuramoto N, Suzuki K, Ueno T, Hada Y, et al. Results of interview about voluntary initiation HAL-Tsukuba. Figshare; 2018. Available from: dx.doi.org/10.6084/m9.figshare.5782881
34. Jeannerod M. Mental imagery in the motor context. Neuropsychologia. 1995;33(11):1419–1432. pmid:8584178
- View Article
- PubMed/NCBI
- Google Scholar
35. Fusco A, Gallotta MC, Iosa M, Morone G, Iasevoli L, Trifoglio D, et al. The dynamic motor imagery of locomotion is task-dependent in patients with stroke. Restor Neurol Neurosci. 2016 Jan 1;34(2):247–56. pmid:26889966
- View Article
- PubMed/NCBI
- Google Scholar
36. Mele AR. Effective Intentions: The Power of Conscious Will. Oxford: Oxford University Press; 2009.
37. Pacherie E. The phenomenology of action: A conceptual framework. Cognition. 2008 Apr;107(1):179–217. pmid:17950720
- View Article
- PubMed/NCBI
- Google Scholar
38. Libet B. Unconscious cerebral initiative and the role of conscious will in voluntary action. Behav Brain Sci. 1985;8:529–66.
- View Article
- Google Scholar
39. Eitam B, Kennedy PM, Higgins ET. Motivation from control. Exp Brain Res. 2013 Sep 1;229(3):475–84. pmid:23288323
- View Article
- PubMed/NCBI
- Google Scholar
40. Mulder T. Motor imagery and action observation: cognitive tools for rehabilitation. J Neural Transm. 2007 Oct;114(10):1265–78. pmid:17579805
- View Article
- PubMed/NCBI
- Google Scholar
41. Gallagher S. The Natural Philosophy of Agency. Philos Compass. 2007 Mar;2(2):347–57.
- View Article
- Google Scholar
42. Bayne T. The Phenomenology of Agency. Philos Compass. 2008 Jan 1;3(1):182–202.
- View Article
- Google Scholar
43. Synofzik M, Vosgerau G, Newen A. Beyond the comparator model: A multifactorial two-step account of agency. Conscious Cogn. 2008 Mar;17(1):219–39. pmid:17482480
- View Article
- PubMed/NCBI
- Google Scholar
44. Zhu J. Intention and Volition. Can J Philos. 2004 Apr 1;34(2):175–93.
- View Article
- Google Scholar
45. Brass M, Haggard P. The What, When, Whether Model of Intentional Action. The Neuroscientist. 2008 Aug 1;14(4):319–25. pmid:18660462
- View Article
- PubMed/NCBI
- Google Scholar
46. McCulloch WS. A heterarchy of values determined by the topology of nervous nets. Bull Math Biophys. 1945 Jun 1;7(2):89–93.
- View Article
- Google Scholar
47. Günther G. Cognition and Volition. A Contribution to a Cybernetic Theory of Subjectivity. In: American Society for Cybernetics, editor. Cybernetics Technique in Brain Research and the Educational Process. Washington D.C.: Alexandria, Va.; 1971. p. 119–35.
48. Latash ML. Synergy. Oxford: Oxford University Press; 2008. xiv+412.
49. Ivanenko YP. Motor Control Programs and Walking. The Neuroscientist. 2006 Aug 1;12(4):339–48. pmid:16840710
- View Article
- PubMed/NCBI
- Google Scholar
50. Ach N. On Volition. Leipzig: Quele & Meyer; 1910.
51. Ach N. Analyse des Willens. Berlin, Wien: Urban & Schwarzenberg; 1935. IX, 460. (Handbuch der biologischen Arbeitsmethoden Abt. 6, Tl E).
52. Kuhl J, Beckmann J. Historical Perspectives in the Study of Action Control. In: Kuhl J, Beckmann J, editors. Action Control: From Cognition to Behavior. Berlin, Heidelberg: Springer; 1985.
53. Grüneberg P, Kadone H, Kuramoto N, Suzuki K. Interview-based questionnaire on robot-assisted voluntary initiation of joint movement. protocols.io; 2018. Available from: dx.doi.org/10.17504/protocols.io.mqnc5ve
54. Chumney D, Nollinger K, Shesko K, Skop K, Spencer M, Newton RA. Ability of Functional Independence Measure to accurately predict functional outcome of stroke-specific population: systematic review. J Rehabil Res Dev. 2010;47(1):17–29. pmid:20437324
- View Article
- PubMed/NCBI
- Google Scholar
55. Bonita R, Beaglehole R. Recovery of motor function after stroke. Stroke. 1988 Dec;19(12):1497–500. pmid:3201508
- View Article
- PubMed/NCBI
- Google Scholar
56. Mahoney FI, Barthel DW. Functional evaluation: the barthel index. Md State Med J. 1965;14:16–65.
- View Article
- Google Scholar

[ref1] 1. Maciejasz P, Eschweiler J, Gerlach-Hahn K, Jansen-Troy A, Leonhardt S. A survey on robotic devices for upper limb rehabilitation. J NeuroEngineering Rehabil. 2014 Jan 9;11(1):3.
View Article
Google Scholar

[2] View Article

[3] Google Scholar

[ref2] 2. Díaz I, Gil JJ, Sánchez E. Lower-Limb Robotic Rehabilitation: Literature Review and Challenges. J Robot. 2011 Nov 17;2011:e759764.
View Article
Google Scholar

[5] View Article

[6] Google Scholar

[ref3] 3. Hogan N, Krebs HI, Rohrer B, Palazzolo JJ, Dipietro L, Fasoli SE, et al. Motions or muscles? Some behavioral factors underlying robotic assistance of motor recovery. J Rehabil Res Dev. 2006;43(5):605. pmid:17123202
View Article
PubMed/NCBI
Google Scholar

[8] View Article

[9] PubMed/NCBI

[10] Google Scholar

[ref4] 4. Belda-Lois J-M, Horno SM, Bermejo-Bosch I, Moreno JC, Pons JL, Farina D, et al. Rehabilitation of gait after stroke: a review towards a top-down approach. J NeuroEngineering Rehabil. 2011 Dec 13;8(1):66.
View Article
Google Scholar

[12] View Article

[13] Google Scholar

[ref5] 5. de Almeida Ribeiro PR, Brasil FL, Witkowski M, Shiman F, Cipriani C, Vitiello N, et al. Controlling Assistive Machines in Paralysis Using Brain Waves and Other Biosignals. Adv Hum-Comp Int. 2013 Jan;2013:3:3–3:3.
View Article
Google Scholar

[15] View Article

[16] Google Scholar

[ref6] 6. Suzuki K, Mito G, Kawamoto H, Hasegawa Y, Sankai Y. Intention-based walking support for paraplegia patients with Robot Suit HAL. In: Miripour B, editor. Climbing and Walking Robots. Rijeka, Shanghai: InTech; 2007. p. 1441–69.

[ref7] 7. Grüneberg P, Kadone H, Suzuki K. Voluntary initiation of movement: multifunctional integration of subjective agency. Front Psychol. 2015;6:688. pmid:26052308
View Article
PubMed/NCBI
Google Scholar

[19] View Article

[20] PubMed/NCBI

[21] Google Scholar

[ref8] 8. Wall A, Borg J, Palmcrantz S. Clinical application of the Hybrid Assistive Limb (HAL) for gait training—a systematic review. Front Syst Neurosci. 2015 Mar 25;9.
View Article
Google Scholar

[23] View Article

[24] Google Scholar

[ref9] 9. Maeshima S, Osawa A, Nishio D, Hirano Y, Takeda K, Kigawa H, et al. Efficacy of a hybrid assistive limb in post-stroke hemiplegic patients: a preliminary report. BMC Neurol. 2011 Dec;11(1).
View Article
Google Scholar

[26] View Article

[27] Google Scholar

[ref10] 10. Ueba T, Hamada O, Ogata T, Inoue T, Shiota E, Sankai Y. Feasibility and Safety of Acute Phase Rehabilitation After Stroke Using the Hybrid Assistive Limb Robot Suit. Neurol Med Chir (Tokyo). 2013;53(5):287–90.
View Article
Google Scholar

[29] View Article

[30] Google Scholar

[ref11] 11. Kawamoto H, Kamibayashi K, Nakata Y, Yamawaki K, Ariyasu R, Sankai Y, et al. Pilot study of locomotion improvement using hybrid assistive limb in chronic stroke patients. BMC Neurol. 2013 Oct 7;13(1):141.
View Article
Google Scholar

[32] View Article

[33] Google Scholar

[ref12] 12. Nilsson A, Vreede KS, Häglund V, Kawamoto H, Sankai Y, Borg J. Gait training early after stroke with a new exoskeleton–the hybrid assistive limb: a study of safety and feasibility. J Neuroengineering Rehabil. 2014;11(1):1.
View Article
Google Scholar

[35] View Article

[36] Google Scholar

[ref13] 13. Watanabe H, Tanaka N, Inuta T, Saitou H, Yanagi H. Locomotion Improvement Using a Hybrid Assistive Limb in Recovery Phase Stroke Patients: A Randomized Controlled Pilot Study. Arch Phys Med Rehabil. 2014 Nov;95(11):2006–12. pmid:25010538
View Article
PubMed/NCBI
Google Scholar

[38] View Article

[39] PubMed/NCBI

[40] Google Scholar

[ref14] 14. Kubota S, Nakata Y, Eguchi K, Kawamoto H, Kamibayashi K, Sakane M, et al. Feasibility of Rehabilitation Training With a Newly Developed Wearable Robot for Patients With Limited Mobility. Arch Phys Med Rehabil. 2013 Jun;94(6):1080–7. pmid:23313372
View Article
PubMed/NCBI
Google Scholar

[42] View Article

[43] PubMed/NCBI

[44] Google Scholar

[ref15] 15. Aach M, Cruciger O, Sczesny-Kaiser M, Höffken O, Meindl RC, Tegenthoff M, et al. Voluntary driven exoskeleton as a new tool for rehabilitation in chronic spinal cord injury: a pilot study. Spine J. 2014 Dec;14(12):2847–53. pmid:24704677
View Article
PubMed/NCBI
Google Scholar

[46] View Article

[47] PubMed/NCBI

[48] Google Scholar

[ref16] 16. Sczesny-Kaiser M, Höffken O, Aach M, Cruciger O, Grasmücke D, Meindl R, et al. HAL® exoskeleton training improves walking parameters and normalizes cortical excitability in primary somatosensory cortex in spinal cord injury patients. J NeuroEngineering Rehabil. 2015 Dec;12(1).
View Article
Google Scholar

[50] View Article

[51] Google Scholar

[ref17] 17. Sakakima H, Ijiri K, Matsuda F, Tominaga H, Biwa T, Yone K, et al. A Newly Developed Robot Suit Hybrid Assistive Limb Facilitated Walking Rehabilitation after Spinal Surgery for Thoracic Ossification of the Posterior Longitudinal Ligament: A Case Report. Case Rep Orthop. 2013 Nov 28;2013:e621405.
View Article
Google Scholar

[53] View Article

[54] Google Scholar

[ref18] 18. Nahmias E, Morris S, Nadelhoffer T. The Phenomenology of Free Will. J Conscious Stud. 2004;11(7–8):162–79.
View Article
Google Scholar

[56] View Article

[57] Google Scholar

[ref19] 19. Pacherie E. Can Conscious Agency Be Saved? Topoi. 2014 Apr 1;33(1):33–45.
View Article
Google Scholar

[59] View Article

[60] Google Scholar

[ref20] 20. Haggard P, Lau H. What is volition? Exp Brain Res. 2013 Sep 1;229(3):285–7. pmid:23727831
View Article
PubMed/NCBI
Google Scholar

[62] View Article

[63] PubMed/NCBI

[64] Google Scholar

[ref21] 21. Götz-Neumann K. Gehen verstehen: Ganganalyse in der Physiotherapie. Stuttgart: Thieme; 2011.

[ref22] 22. Zhu J. The Conative Mind: Volition and Action. Saarbrücken: LAP LAMBERT Academic Publishing; 2012. 140 p.

[ref23] 23. Ach N. Über die Willenstätigkeit und das Denken. Eine experimentelle Untersuchung mit einem Anhange: Über das Hippsche Chronoskop. Göttingen: Vandenhoeck & Ruprecht; 1905.

[ref24] 24. Pfänder A. Phänomenologie des Wollens. Eine psychologische Analyse. Leipzig: J. A. Barth; 1900.

[ref25] 25. Lindworsky J. Der Wille: seine Erscheinung und seine Beherrschung nach den Ergebnissen der experimentellen Forschung. Leipzig: J.A. Barth; 1921.

[ref26] 26. Michotte A, Prum E. Le Choix Volontaire et ses antededents immediats. Arch Psychol. 1910;38–39:113–296.
View Article
Google Scholar

[71] View Article

[72] Google Scholar

[ref27] 27. Synofzik M, Vosgerau G. Beyond the comparator model. Conscious Cogn. 2012 Mar;21(1):1–3. pmid:22301453
View Article
PubMed/NCBI
Google Scholar

[74] View Article

[75] PubMed/NCBI

[76] Google Scholar

[ref28] 28. Chambon V, Sidarus N, Haggard P. From action intentions to action effects: how does the sense of agency come about? Front Hum Neurosci. 2014;8:320. pmid:24860486
View Article
PubMed/NCBI
Google Scholar

[78] View Article

[79] PubMed/NCBI

[80] Google Scholar

[ref29] 29. Kühn S, Brass M, Haggard P. Feeling in control: Neural correlates of experience of agency. Cortex. 2013 Jul;49(7):1935–42. pmid:23067726
View Article
PubMed/NCBI
Google Scholar

[82] View Article

[83] PubMed/NCBI

[84] Google Scholar

[ref30] 30. Metzinger T. Conscious Volition and Mental Representation: Toward a More Fine-Grained Analysis. In: Sebanz N, Prinz W, editors. Disorders of Volition. Cambridge, Mass.: MIT Press; 2006.

[ref31] 31. Horgan T, Timmons M. Introspection and the Phenomenology of Free Will: Problems and Prospects. J Conscious Stud. 2011;18(1):180–205.
View Article
Google Scholar

[87] View Article

[88] Google Scholar

[ref32] 32. Grüneberg P, Kadone H, Kuramoto N, Suzuki K, Sankai Y. Voluntary initiation questionnaire JP-ENG. Figshare; 2018. Available from: dx.doi.org/10.6084/m9.figshare.5782812

[ref33] 33. Grüneberg P, Kadone H, Kuramoto N, Suzuki K, Ueno T, Hada Y, et al. Results of interview about voluntary initiation HAL-Tsukuba. Figshare; 2018. Available from: dx.doi.org/10.6084/m9.figshare.5782881

[ref34] 34. Jeannerod M. Mental imagery in the motor context. Neuropsychologia. 1995;33(11):1419–1432. pmid:8584178
View Article
PubMed/NCBI
Google Scholar

[92] View Article

[93] PubMed/NCBI

[94] Google Scholar

[ref35] 35. Fusco A, Gallotta MC, Iosa M, Morone G, Iasevoli L, Trifoglio D, et al. The dynamic motor imagery of locomotion is task-dependent in patients with stroke. Restor Neurol Neurosci. 2016 Jan 1;34(2):247–56. pmid:26889966
View Article
PubMed/NCBI
Google Scholar

[96] View Article

[97] PubMed/NCBI

[98] Google Scholar

[ref36] 36. Mele AR. Effective Intentions: The Power of Conscious Will. Oxford: Oxford University Press; 2009.

[ref37] 37. Pacherie E. The phenomenology of action: A conceptual framework. Cognition. 2008 Apr;107(1):179–217. pmid:17950720
View Article
PubMed/NCBI
Google Scholar

[101] View Article

[102] PubMed/NCBI

[103] Google Scholar

[ref38] 38. Libet B. Unconscious cerebral initiative and the role of conscious will in voluntary action. Behav Brain Sci. 1985;8:529–66.
View Article
Google Scholar

[105] View Article

[106] Google Scholar

[ref39] 39. Eitam B, Kennedy PM, Higgins ET. Motivation from control. Exp Brain Res. 2013 Sep 1;229(3):475–84. pmid:23288323
View Article
PubMed/NCBI
Google Scholar

[108] View Article

[109] PubMed/NCBI

[110] Google Scholar

[ref40] 40. Mulder T. Motor imagery and action observation: cognitive tools for rehabilitation. J Neural Transm. 2007 Oct;114(10):1265–78. pmid:17579805
View Article
PubMed/NCBI
Google Scholar

[112] View Article

[113] PubMed/NCBI

[114] Google Scholar

[ref41] 41. Gallagher S. The Natural Philosophy of Agency. Philos Compass. 2007 Mar;2(2):347–57.
View Article
Google Scholar

[116] View Article

[117] Google Scholar

[ref42] 42. Bayne T. The Phenomenology of Agency. Philos Compass. 2008 Jan 1;3(1):182–202.
View Article
Google Scholar

[119] View Article

[120] Google Scholar

[ref43] 43. Synofzik M, Vosgerau G, Newen A. Beyond the comparator model: A multifactorial two-step account of agency. Conscious Cogn. 2008 Mar;17(1):219–39. pmid:17482480
View Article
PubMed/NCBI
Google Scholar

[122] View Article

[123] PubMed/NCBI

[124] Google Scholar

[ref44] 44. Zhu J. Intention and Volition. Can J Philos. 2004 Apr 1;34(2):175–93.
View Article
Google Scholar

[126] View Article

[127] Google Scholar

[ref45] 45. Brass M, Haggard P. The What, When, Whether Model of Intentional Action. The Neuroscientist. 2008 Aug 1;14(4):319–25. pmid:18660462
View Article
PubMed/NCBI
Google Scholar

[129] View Article

[130] PubMed/NCBI

[131] Google Scholar

[ref46] 46. McCulloch WS. A heterarchy of values determined by the topology of nervous nets. Bull Math Biophys. 1945 Jun 1;7(2):89–93.
View Article
Google Scholar

[133] View Article

[134] Google Scholar

[ref47] 47. Günther G. Cognition and Volition. A Contribution to a Cybernetic Theory of Subjectivity. In: American Society for Cybernetics, editor. Cybernetics Technique in Brain Research and the Educational Process. Washington D.C.: Alexandria, Va.; 1971. p. 119–35.

[ref48] 48. Latash ML. Synergy. Oxford: Oxford University Press; 2008. xiv+412.

[ref49] 49. Ivanenko YP. Motor Control Programs and Walking. The Neuroscientist. 2006 Aug 1;12(4):339–48. pmid:16840710
View Article
PubMed/NCBI
Google Scholar

[138] View Article

[139] PubMed/NCBI

[140] Google Scholar

[ref50] 50. Ach N. On Volition. Leipzig: Quele & Meyer; 1910.

[ref51] 51. Ach N. Analyse des Willens. Berlin, Wien: Urban & Schwarzenberg; 1935. IX, 460. (Handbuch der biologischen Arbeitsmethoden Abt. 6, Tl E).

[ref52] 52. Kuhl J, Beckmann J. Historical Perspectives in the Study of Action Control. In: Kuhl J, Beckmann J, editors. Action Control: From Cognition to Behavior. Berlin, Heidelberg: Springer; 1985.

[ref53] 53. Grüneberg P, Kadone H, Kuramoto N, Suzuki K. Interview-based questionnaire on robot-assisted voluntary initiation of joint movement. protocols.io; 2018. Available from: dx.doi.org/10.17504/protocols.io.mqnc5ve

[ref54] 54. Chumney D, Nollinger K, Shesko K, Skop K, Spencer M, Newton RA. Ability of Functional Independence Measure to accurately predict functional outcome of stroke-specific population: systematic review. J Rehabil Res Dev. 2010;47(1):17–29. pmid:20437324
View Article
PubMed/NCBI
Google Scholar

[146] View Article

[147] PubMed/NCBI

[148] Google Scholar

[ref55] 55. Bonita R, Beaglehole R. Recovery of motor function after stroke. Stroke. 1988 Dec;19(12):1497–500. pmid:3201508
View Article
PubMed/NCBI
Google Scholar

[150] View Article

[151] PubMed/NCBI

[152] Google Scholar

[ref56] 56. Mahoney FI, Barthel DW. Functional evaluation: the barthel index. Md State Med J. 1965;14:16–65.
View Article
Google Scholar

[154] View Article

[155] Google Scholar

Figures

Abstract

Introduction

Results

Voluntary initiation as phenomenal performance

Voluntary initiation as a distinct act in the system of action control

Two groups of robot-assisted voluntary initiation

Discussion

Subjective evidence for voluntary initiation of joint movement

Efficacious action consciousness

Heterarchic organization of action control

Clinical implications

Materials and methods

HAL therapy session

Experimental procedure

Interview-based questionnaire

Subjects and statistical data

Clinical assessment

References