Motor imagery of body movements that can’t be executed on Earth

1 Introduction

There is ample evidence that spatial orientation on Earth relies – among others, on the gravitational vertical as an omnipresent, stable and intuitively accessible reference [36]. This reference is absent during spaceflight, which challenges astronauts’ spatial orientation: With their eyes open, they may feel upside down in an upside down environment, right-side up in an upside down environment, or upside down in a right-side up environment [24, 31]. When closing their eyes, they may feel right side up, upside down, or lose all sense of orientation in space [15]. These visual orientation illusions vary in dependence on visual, tactile, and cognitive cues [23]. Since such illusions are known to provoke space motion sickness [28, 31] and to degrade cognitive performance [16], their recurrence could adversely affect the success of a space mission.

Astronauts are typically prepared for the sensory, motor and cognitive challenges of weightlessness through training under water and on parabolic flights. A quick, simple and inexpensive alternative could be motor imagery (MI) [2]. MI is a cognitive process during which the representation of a specific action is internally reproduced in working memory from a first-person perspective, without any overt motor output [9]. The mental reproduction typically refers to the kinesthetic and/or visual sensory modality [1], and is thought to rely on similar action representations as actual movements are [20]. Similarities of MI and motor execution (ME) are represented in shared neural networks [30], responses of the autonomic nervous system [10], and – at the behavioral level – shared temporal structures [17]. Although research suggests a partial overlap of neural substrates, more precisely of motor and sensory regions (primary and secondary motor cortices, posterior parietal cortex, basal ganglia, and cerebellum; [20, 30]), the neural activation for the imagined movements seems to be less intense and localized compared to physical performance of the task [13].

As the motor representation contains all elements of the movement, a well-developed motor representation leads to accurate MI and thus increases the amount of functional equivalence between MI and ME [19]. Just as repeated movements in physical exercise, modify neural processes and thus refine future motor commands, repeated MI can also refine motor representations [14]. Such mental practice is particularly effective when all sensory, motor and affective characteristics are closely similar to the physical activity being trained [19].

To address the multifaceted nature of motor imagery, one has to examine different key characteristics of the motor imagery ability: generation (formation of a mental image without the presence of external visual stimuli), visual and kinesthetic imagery ability (clarity and intensity of the mental representation), controllability (manipulating mental representations) and the temporal organization [3]. In the last decades a variety of behavioral methods has been established to test for the accuracy of motor representations [7]. Among them are the Movement Imagery Questionnaire and the Kinesthetic and Visual Imagery Questionnaire [27], which assesses subjects’ vividness of visual and kinesthetic MI through self-report. It has been shown that the task requirements (such as changing directions or dual-task) in walking are reflected in the subject's subjective rating of MI [21].

Another assessment tool is the Controllability of Motor Imagery Test [33]. Subjects generate and manipulate imagined postures of body parts in response to verbal instructions, and afterwards assume the final imagined body position. Recent studies have shown that young as well as older adults (up to the age of 70 years of age) are able to largely regenerate the instructed body positions [21, 35]. However, it remains unclear whether they are also able to imagine movements that are impossible to perform under the presence of gravity, such as body segment movements while free-floating and whole-body movements about the pitch and roll axis. Such movements are characteristic for activities in weightlessness, and the ability to imagine them is therefore an important prerequisite for astronaut preflight training based on MI.

Previous work has shown that humans can imagine a whole-body rotation about the pitch and roll axis, which is impossible to physically execute on Earth; however, speed and accuracy is lower than for a whole-body rotation about the yaw axis, which is common on Earth [4, 8]. The present study expands this work from single whole-body rotations to sequences of whole-body and body-segment rotations. If the ability to imagine movements that are impossible on Earth should extend beyond single motor acts and include more complex behavioural sequences, this would encourage us to further pursue MI as a possible training approach for astronauts.

2 Methods

2.2 Procedures

Subjects were examined with a modified version of the Controllability of Motor Imagery (CMI) test. In the original CMI test [33], subjects stand with their eyes closed on a cross marked on the floor and receive six consecutive instructions for moving a body segment into a specific position (e.g. “take a step with your right foot to the right side”). They are asked to remain physically still and only to imagine that they execute the requested movement. After the sixth instruction, subjects open their eyes and physically assume the final imagined body posture. Performance on each trial is quantified by awarding one point if final body posture is correct, and by registering the time needed to assume the final configuration (as measured with an electronic stopwatch). The test consists of ten trials; overall performance is calculated as the sum of all points (“CMI score”) and the mean time of correct trials (“Time”).

In the modified CMI test used in the present study, subjects didn’t physically assume the final imagined body posture but rather demonstrated it on a segmented manikin of 20 cm length, which they held within a reference volume consisting of a “floor” and two adjacent “walls”, all with 25 cm side length. This modification became necessary because in some conditions of our study it was impossible for subjects to assume the final position themselves due to gravitational constraints. In condition CMI_ground, subjects received the same instructions for moving their body segments as in the original CMI test. In condition CMI_float1, subjects were asked before each instruction to imagine that they float above the floor. This was followed by instructions as in the original CMI test, except that one of the instructions was to rotate not a body segment but rather the whole body (e.g. rotate around longitudinal, transversal, or sagital axis). Condition CMI_float2 differed only by using two rather than one instruction to rotate the whole body. Each condition consisted of ten trials. Note that in both CMI_float conditions, subjects were unable to assume the final body posture themselves because of gravitational constraints; hence the need to use a segmented manikin (see Fig. 1).

OPEN IN VIEWER

Fig.1

Schematic example for position of adjusted manikin in CMI_ground (left) and CMI_float (right).

Subjects were randomly assigned to two experimental groups. One was given the conditions CMI_ground and CMI_float1, while the other was given the conditions CMI_ground and CMI_float2. This allowed us to calculate within-subject comparisons of CMI on ground and afloat and between-subject comparisons of CMI with one and with two whole-body instructions. We decided not to administer all three conditions, CMI_ground, CMI_float1 and CMI_float2, to each subject, to limit the effects of fatigue. The order of conditions CMI_ground and CMI_float was balanced in each group. A rest break of five minutes was provided between the two conditions.

After completing both CMI conditions, subjects rated their subjective vividness of visual and kinesthetic imagery using the scale of the Kinesthetic and Visual Imagery Questionnaire (KVIQ) [19]. This instrument uses a five-point Likert scale for the vividness of visual imagery (1 = no image to 5 = image as clear as really seen), and a second one for the vividness of kinesthetic imagery (1 = no feeling to 5 = feeling as if really executed).

3 Results

As Table 1 illustrates, subjects achieved higher CMI scores and shorter Times in CMI_ground than in CMI_float. ANOVA accordingly revealed a significant effect of Condition for CMI scores (F (1, 66) = 82.736; P < 0.001; η² = 0.56) and for Time (F (1, 66) = 36.575; p < 0.001; η² = 0.37). The effect of Group was non-significant for either variable. No significant main effects or interactions of Group and Gender emerged for either variable.

Further from Table 1, vividness ratings were higher in Ground than in Float. ANOVA confirmed significance of Condition for visual vividness (F (1, 66) = 28.407; P < 0.001; η² = 0.30) as well as for kinesthetic vividness (F (1, 66) = 32.427; p < 0.001; η² = 0.33); all other main effects and interactions were non-significant.

Stepwise multiple linear regression showed a significant relationship between regressors and CMI scores of condition CMI_float (F(4,59) = 7.674; R² = 0.34; P < 0.001). As Table 2 illustrates, CMI scores of condition CMI_float are associated with CMI scores of condition CMI_ground, with Time of condition CMI_float, and with visual vividness ratings.

The purpose of the present study was to investigate the controllability of MI of movements that are impossible to perform under the presence of gravity by instructing to imagine moving body segments and self-rotations.

The results of the condition comparison for CMI show that subjects perform less accurately in the CMI_float condition, suggesting that impossible movements in a floating position are more difficult to imagine. As no significant Group x Condition occurred it can be concluded that the increasing complexity due the floating movements is independent of the amount of body rotations. Although we could show that accuracy of MI declines when adding instructions for self-rotations we are led to the conclusion that self-rotations along body axis are possible to imagine despite the conflict of the gravitational reference, as it was reported in previous studies [4, 8]. The observed decline in MI with body rotations might be due to the fact that a combination of rotations and segment movements were demanded in the present study. The fact that subjects needed less time to adjust the manikin is further evidence, that the imagined position (floating or on ground) has an influence on MI performance. As the procedure indicating the final configuration was different as in the previous studies using the CMI test the validity of the results remain unclear. However, the mean scores of CMI_ground were in a comparable range as in the mentioned studies [21, 32]. It remains unclear, if astronauts or participants that have experienced weightlessness would perform better in the CMI_float condition as they have motor experiences with such movements and therefore draw on existing mental representation. This issue deserves further empirical studies and could be realized by comparing CMI performance before and after a first parabolic flight.

Condition differences were also detected concerning the subjective rating of MI performance with higher clarity for CMI_ground. Condition differences for subjective ratings have been previously reported for imagined walking with different task requirements [22] and are further evidence that the specific affordance of a movement sequence is reflected in perception of MI accuracy. However, it should be further noted that at least the visual clarity for CMI_float is in a comparable range to young adults performing the original CMItest [22].

The multiple linear regression analysis showed that the CMI_float performance is in significant positive relation to the CMI_ground performance and the subjective visual MI performance. This result suggests that the CMI_float performance is dependent of the individual controllability of MI and the ability to imagine the visual aspects of the movements accurately. These findings could be interpreted as further evidence that it is possible to imagine floating movements but the performance is related to the individual MI abilities. The time for correct trials and CMI_float performance show a negative relation, which means that the quicker the response the better the MI performance. This result is in accordance with a previous study [32] and suggests that the response time could be seen as a reference for a good motor representation or the ability to maintain the mental image. Thereby it should be investigated if the CMI performance could be improved by training and moreover if specific moderating factors such as working memory or experiences are prerequisite for learning effects [32] In a recent study from our lab, we were able to show, that subjects with high scores on the CMI performed on almost the same level compared to a physical training group (Schott, Schneider, & Hohmann, in prep).

Since MI only involves cognitive processes, it can be practiced or implemented in various settings for several purposes. Besides its usage for athletes or surgeons for skill acquisitions or preparation of performance [6], MI can be especially beneficial in patients with neurologic, orthopedic, or vestibular impairments. Due to benefits on the behavioral and the neural level, MI has been already widely used in the therapeutic domain for instance for (re-)learning movement skills after stroke [26], to prevent functional loss during immobilization after wrist fractures [34], to improve postural control in older adults [18], or to support patient with bilateral vestibular impairments in action planning [12]. The modified version of the CMI could especially be used in these settings were the physical reproduction of MI is inconvertible.

The results of to present study give further evidence for the feasibility and benefits of MI in testing of mental representation but also improving of motor performance and neural circuity. This notion includes target population with specific requirements, such as astronauts, patients or older adults. However, to explain the underlying mechanism and efficiency of MI to prepare and as for space missions and as countermeasure for space induced alterations further research is mandatory. It has been shown that a long duration spaceflight initiates cortical reorganization with alterations in vestibular and motor-related regions [11]. These alterations can worsen the vestibular function at re-entry after space flights. MI of vestibular challenging movements and postures, such as in the CMI, could be a useful and beneficial countermeasure for astronauts. Thereby MI could be used to prepare for the motor challenges after re-entry but also maintain cortical organization and function and thus health during space missions with long exposure to microgravity. A four week practice with MI can completely prevent the prolongation of corticospinal inhibition after wrist-hand immobilization [5] and 1-week MI of hand-movement can increase hemodynamic response of the primary and secondary motor cortex and the cerebellum [25]. Moreover, MI could be used during space flight to prepare for daily activities after landing. This could be realized by MI scripts of movement sequences with focus on gravity when walking or balancing. Finally, MI could be used on the ISS for planning of action such as floating from module A to module B.

The results of the present study are also of practical relevance. First, it is possible to imagine movements which are impossible to physically perform on earth, although it is more difficult to perform accurately. Thus the prerequisite for mental training is fulfilled to prepare astronauts for the in the beginning described challenges for spatial orientation [24, 31]. Second, as no Gender effect could be found, mental practice to increase motor performance in weightlessness could be also beneficial for female astronauts. Third, the modified version of the CMI seems to be useful method to test for MI accuracy and thus could be also used in other settings were the physical reproduction of MI is inconvertible such as in weightlessness or diving.

5 Conclusion

Based on the present study we can conclude that it is possible to perform MI of movements that cannot be executed on earth. Although subjects show higher performance for movements which can be executed on earth, we are likely to the conclusion that the prerequisite for mental training is fulfilled and astronauts could benefit by using MI in preflight training.