Skip to main content
NCBI home page
Human Brain Mapping logoLink to Human Brain Mapping
. 2008 Apr 15;30(3):990–997. doi: 10.1002/hbm.20568

A new vibrator to stimulate muscle proprioceptors in fMRI

Marie Montant 1,2,✉, Patricia Romaiguère 1,3, Jean‐Pierre Roll 1,3
PMCID: PMC6871194  PMID: 18412129

Abstract

Studying cognitive brain functions by functional magnetic resonance imaging (fMRI) requires appropriate stimulation devices that do not interfere with the magnetic fields. Since the emergence of fMRI in the 90s, a number of stimulation devices have been developed for the visual and auditory modalities. Only few devices, however, have been developed for the somesthesic modality. Here, we present a vibration device for studying somesthesia that is compatible with high magnetic field environments and that can be used in fMRI machines. This device consists of a poly vinyl chloride (PVC) vibrator containing a wind turbine and of a pneumatic apparatus that controls 1–6 vibrators simultaneously. Just like classical electromagnetic vibrators, our device stimulates muscle mechanoreceptors (muscle spindles) and generates reliable illusions of movement. We provide the fMRI compatibility data (phantom test), the calibration curve (vibration frequency as a function of air flow), as well as the results of a kinesthetic test (perceived speed of the illusory movement as a function of vibration frequency). This device was used successfully in several brain imaging studies using both fMRI and magnetoencephalography. Hum Brain Mapp, 2009. © 2008 Wiley‐Liss, Inc.

Keywords: proprioception, somesthesic mechanosensitivity, fMRI compatibility, pneumatic stimulator, tendon vibration, illusion of movement

INTRODUCTION

Muscle proprioception plays a major role in both somesthesia and kinesthesia [Gilhodes et al., 1986; Goodwin et al., 1972; Matthews, 1982; Roll et al., 1996; Roll and Vedel, 1982]. Muscle proprioception originates from the activity of stretch receptors located in the muscles (i.e., the muscle spindles). These receptors are particularly sensitive to the lengthening of their bearing muscles; they code for both muscle length and speed of muscle lengthening. During movements, muscle proprioceptive feedback is associated with cutaneous feedback and with information about the intended movement generated by the motor command itself. Thus, to study human muscle proprioception in isolation, one has to dissociate muscle sensory information from motor commands and tactile feedback. One way is to apply mechanical vibration (0.2–0.5 mm in amplitude) to muscle tendons [see Roll et al., 1989]. In the absence of visual feedback of the vibrated limb, tendon vibration induces an illusory sensation of movement [Eklund, 1972; Goodwin et al., 1972; Roll and Gilhodes, 1995; Roll and Vedel, 1982], whose direction is that of the movement that would have stretched the vibrated muscle. These kinesthetic illusions are generated by the stimulation, through vibration, of primary endings of muscle spindles [Burke et al., 1976a, b; Roll et al., 1989; Roll and Vedel, 1982]. There is a quantitative relationship between the frequency of vibration and the perceived velocity of the illusory movements. This perceived velocity peaks for vibration between 70 and 100 Hz [Jones, 1988; Roll and Vedel, 1982].

The study of muscle proprioception has benefited from a substantial research since the 70s, through both behavioral and electrophysiological methods. Although muscle proprioception was first considered as a second order sensitivity, mainly operating at spinal and unconscious levels, it has become clear that it does access consciousness, and thus plays a decisive part in perceptual and cognitive functions.

It is essential to pursue research in the field of muscle proprioception at least for two reasons. First, muscle proprioception provides a frame of reference for most of the other sensory modalities by providing continuous information about the position of the body and body parts, including those that hold the sensory organs (i.e., head, eyes, hands…). This frame is necessary for the coherent perception of environment [Roll, 1998]. Experimental manipulations of muscle proprioception can generate perceptual distortions in several sensory modalities, including vision, a dominant sensory modality in humans. For example, these manipulations can generate the illusion that a visual target is moving even though the visual scene on the retina remains static [Roll et al., 1991]. Second, in cognitive skills like handwriting, muscle afferent information may be critical for the concatenation of simple motor programs into a fully formed motor plan [Marsden, 1982]. Indeed, a case study on a deafferented patient [Teasdale et al., 1993] suggested that, in absence of vision, proprioceptive information is crucial for the spatial calibration of handwriting movements.

Although fMRI is of particular interest for the study of proprioception, and more widely somesthesia, it is difficult to run such experiments in fMRI machines because of the technical constraints of high strength magnetic fields. For example, ferromagnetic objects are attracted by the magnet, and most electric devices do not function properly in a high magnetic field. Technical solutions have been found to avoid these problems in areas such as visual perception. Less effort has gone into developing such solutions for the study of somesthesia.

For the study of somesthesia in fMRI, two main types of vibrators have been proposed: piezoelectric [e.g., Harrington et al., 2000] and magnetomechanical vibrotactile devices [MVDs, e.g., Graham et al., 2001]. Although piezoelectric devices are appropriate for superficial tactile stimulation in terms of amplitude (3.39 μm/V) and frequency (1–300 Hz frequency range), their main drawback is that the amplitude of the produced vibration is too small (less than 300 μm) for triggering robust muscle spindle stimulation. Gizewski et al., [ 2005] proposed a new device with greater vibration amplitude (up to 1 mm). This device, however, drives only one vibrator, which means that only one muscle or one group of adjacent muscles can be stimulated at once. Given the size of the fixation arm and vibrator (in Fig. 1 of the above cited article, the total length of the device approximates the actual length of the subject's arm), it would be difficult to design and fit a flexible device with multiple piezoelectric vibrators of this type in a scanner; yet it is now well established that several vibrators are required to generate complex and accurate patterns of vibration giving rise to multidimensional illusory movements [Albert et al., 2006; Roll et al., 2004].

According to Graham et al., [ 2001], MVDs are able to efficiently stimulate muscle proprioceptors in a fMRI machine. However, because these vibrators use the static magnetic field of the fMRI scanner and the Lorenz force mechanism, the parameters of their vibration change with the position and orientation of these stimulators within the magnetic field. As a consequence, the parameters of the stimulation may vary for each subject and each muscle.

A third type of fMRI‐compatible device, driven by constant air pulse, has been used recently for mapping the somatosensory cortex [Briggs et al., 2004; Gelnar et al., 1998; Stippich et al., 1999]. In these studies, the pneumatic stimulators seem to be more appropriate for investigating tactile rather than muscle proprioception since they generate only low frequency vibration [e.g., 4 Hz in Stippich et al., 1999], whereas illusions of movement are obtained with vibration frequencies from 40 to 120 Hz [Roll and Vedel, 1982]. The pneumatic device proposed by Briggs et al., [ 2004] can produce vibration at up to 100 Hz. However, as the authors acknowledge, “[at high frequencies], a shorter air pulse duration is required to prevent bursting the latex diaphragm” (p. 641), for example, 3 s for 8–20 Hz vibration frequencies. Therefore, we can assume that for vibrations around 80 Hz (which is the optimal frequency for evoking illusions of movement), even shorter air pulse durations have to be set. Because very short tendon vibrations do not allow the full development of movement illusion, it is unlikely that this device is appropriate for the study of muscle proprioception.

In this article, we present a new fMRI‐compatible and flexible pneumatic device that is able to activate muscle spindles from several muscles and to generate reliable and complex illusions of movements of any body part.

MATERIALS AND EQUIPMENT

Vibrator Prototype

The body of the vibrator is a cylinder that is made of poly vinyl chloride (PVC), a self‐lubricating material; it contains a PVC wind turbine to which an off‐centered mass is attached (see Fig. 1a,b). The off‐centered mass is made of barite or barium sulfate (BaSO4), a nonferrous mineral with a volume magnetic susceptibility close to zero (χ = −1.269 × 10−6) The susceptibility difference between barite and water is −0.765 ppm and the electrical conductivity of barite is 147 μS/cm (for comparison, the electrical conductivity of copper is 596 × 109 μS/cm). Barite was chosen for the off‐centered mass because it is almost amagnetic and heavy, with a density of 4.5 g/cm3 (for comparison, the density of charcoal is 0.2 g/cm3, that of limestone is 2 g/cm3, and that of nickel silver is 8.4 g/cm3). The rotation of the off‐centered mass generates tangential vibrations transmitted to the vibrator body. The vibration frequency and amplitude depend on the angular velocity of the rotor, which is proportional to the air inflow. To attenuate the acoustic noise generated by the air outflow, the vibrator is equipped with a silencer. Using a silencer is necessary in particular during the training of the subjects outside the scanning room. Four lugs are placed on the external face of the cylinder to allow the experimenter to attach the vibrator to someone's arm or leg using rubber bands (see Fig. 2). Rubber was used because its elasticity prevents the transmission of vibration to the tendons of adjacent and antagonistic muscle groups.

Figure 1.

Figure 1

Open in a new tab

Section schematics of an fMRI‐compatible pneumatic vibrator prototype. Top view (Panel a), side view (Panel b). The top view is an horizontal section going through the upper part of the vibrator presented in Panel b (see bidirectional arrow). All PVC parts appear in grey color. Note that the cylinder that contains the turbine has thick lateral walls (Panel a). In this prototype, the barite weighs 1.81 g while the turbine weighs 5.49 g. The eccentricity of the barite is 5 mm from the rotation axis. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

Figure 2.

Figure 2

Open in a new tab

Pneumatic vibrators attached to a subject's wrist. Here, the mechanical vibration is applied to the abductors and adductors of the hand. The vibrators are strapped together using rubber bands.

Pneumatic Apparatus

The pneumatic vibrators do not use any electrical power inside the scanning room. The wind turbines of the vibrators are moved with compressed air. The device is controlled electronically by an apparatus (described below) that is placed outside the scanning room.

Six bars (600 kPa) of compressed air generated by a DeVilbiss LT100 HP3 compressor go through six electrogates that work independently (see Fig. 3). The electrogates function on a binary mode (i.e., open or shut). From the electrogates, the air is sent to the vibrators through six flowmetric gates that regulate the air flow independently by means of six endless screws controlled by six DC motors. This setup makes it possible to use up to six vibrators in parallel with specific vibration patterns for each. The aperture of the flowmetric gates goes from 0 to 100% in 1% increments. The higher the air flow, the higher the frequency of vibration. Each flowmetric gate is connected to only one vibrator, through 15 m of flexible polyethylene tubes. The electro and flowmetric gates are controlled by a PC using LabVIEW 7.1 (National Instruments). Labview functions are also used to synchronize the stimulation device with the fMRI machine. The time lag between the computer command and the first vibration is 101.63 ± 6.2 ms (this was estimated from the signal of a piezoelectric shock detector that we strapped onto a vibrator, see next section). A delay of ∼1 s is necessary before the vibration reaches a plateau in amplitude and frequency (see Fig 4a, Panel A).

Figure 3.

Figure 3

Open in a new tab

Pneumatic apparatus including six flowmetric gates that can be activated simultaneously. The pattern of stimulation is set independently for each pneumatic electrogate.

Figure 4.

Figure 4

Open in a new tab

(a) Signal obtained from the piezoelectric shock detector with a 60% aperture of the flowmetric gate (Panel A) and the resulting FFT (Panel B). (b) Mean frequency of vibration (and standard errors) as a function of the percentage of aperture of the flowmetric gate, inside and outside the scanner. Note that error bars are mostly smaller than the size of the symbols. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

Note that our pneumatic device is fMRI safe. Devices that are potentially dangerous for participants or experimenters usually include ferromagnetic components or electrically conductive devices. Our device contains no electrically conductive element or any ferromagnetic component. Barite is particularly safe since it is used in suspension as an oral MRI and X‐ray contrast agent for examination of the human gastrointestinal tract [Courbière et al., 2005; Li et al., 1991].

DEVICE TESTING PROCEDURE

First, we tested the fMRI compatibility of the vibrator to make sure it does not interfere with fMRI data. This includes a series of calibration measures inside and outside the scanner. Second, we ran a kinesthetic test to show that this tool can be used efficiently for the study of muscle proprioception.

fMRI Compatibility

To achieve fMRI compatibility, three conditions have to be met: (1) the device must be fMRI safe, (2) the device should not compromise image quality, and (3) the device must operate as intended in the fMRI environment. The first issue was addressed above in the Materials and equipment section. As concerns the second issue, some nonferrous materials are not compatible with the fMRI environment. They can produce inhomogeneities in the static magnetic field that will distort the images [Schenck, 1996]. The fMRI compatibility of any new device has to be assessed before using it. To test for the fMRI compatibility of our vibrator prototype, we ran an EPI sequence (36 slices, 192 mm FOV, 64 matrix, 3 s TR, 90° angle of impulse) with a “phantom” in a 3T Bruker magnet (MedSpec 30/80 Avance). A shielded bird cage quadrature head coil was used for transmission and reception. The operating system was ParaVision 2.1. The phantom consisted of a sphere of glass filled with water, copper sulfate and salt (1 g/l). The electrical conductivity of the phantom was analogous to the electrical conductivity of a human head (same tuning and matching of the head coil). Four vibrators were placed in the machine, about 20 cm from the phantom. We did not put the vibrators directly on the phantom for two reasons. First, in real experiments, it would be very difficult to vibrate the tendons of head muscles (neck or eye muscles) because of the proximity of the antenna and the padded table. Second, directly vibrating the phantom will distort the magnetic field because any movement of the phantom produces distortions. For the same reason, vibrating the head of a participant would generate artifacts in the images. As concerns the third issue, to test the efficiency and reliability of the vibrators, we recorded inside and outside the scanner the vibration frequency of the vibrator as a function of the percentage of aperture of the flowmetric gate (frequency calibration). The parameters of these calibration measurements are the following.

The frequency of the vibration depends on air pressure, the percentage of aperture of the flowmetric gate, and the length and diameter of the polyethylene tubes that connect the flowmetric gate to the vibrators. In the present study, we used an air pressure of 6 bars (600 kPa). The tubes were 15 m long with an internal diameter of 6.8 mm. We obtained two series of measures of the vibration frequency as a function of the percentage of aperture of the flowmetric gate: one with the vibrator in the center of a 3T fMRI machine and one with the vibrator in the control room (∼10 m away from the machine). For these calibration measures, we strapped a vibrator onto an fMRI‐compatible piezoelectric shock detector (Murata PKS1‐4A1, Radiospare). The program driving the vibrator was written in LabVIEW 7.1. The vibration signal was acquired using a National Instruments analog input device (PXI 6289). Specific acquisition and analysis functions were written in LabVIEW 7.1. Acquisition runs of 4 s were recorded at 5000 Hz and synchronized with the beginning of the vibration. Each run was processed independently. Fast Fourier Transform (FFT) was computed to drive the frequency analysis. For each run, the estimated vibration frequency corresponds to the frequency with the highest amplitude (Fig. 4a shows the shock detector signal and the corresponding FFT for a representative trial). We repeated the measurement five times for each aperture of the flowmetric gate (10–100%). Note that the exact same apparatus was used for the measures inside and outside the scanner (i.e., same vibrator, same tubing, same pneumatic control devices and same shock detector).

Kinesthetic Test

Previous kinesthetic data obtained with classical electromagnetic vibrators showed that there is a robust relationship between vibration frequency and the perceived speed of illusory movements [Roll and Vedel, 1982; Vedel and Roll, 1983]. Typically, the perceived velocity of the illusory movement follows a parabolic function that peaks around 80 Hz [e.g., Roll and Vedel, 1982]. We attempted to replicate this relationship by using the pneumatic vibrator. We applied increasing vibration frequencies (40–100 Hz) to the tendon of the extensor muscle of the left wrist of six right‐handed participants. Participants were holding a pencil in their right hand and were instructed to track, on a graphic table, the illusory movement perceived by the left hand. The amplitude and the velocity of the right hand (tracking) movement were recorded. The order of the vibration frequencies was counterbalanced across participants. This test was not done in the scanner because the graphic table is not fMRI compatible.

All participants gave informed consent. The experiment was approved by the local ethic committee

RESULTS

fMRI Compatibility

We measured the signal/noise ratio, that is, the mean fMRI signal of the phantom versus the standard deviation of the fMRI signal outside the phantom. Activating the vibrators had no impact on the signal/noise ratio (all P values > 0.10), independently of the percentage of aperture of the flowmetric gates (50 or 100%) and independently of the number of vibrators activated (1 or 4). The mean signal/noise ratio was 61.25 (±3.99) with the four vibrators activated (100% aperture of the flowmetric gates, 21 measures) and 62.03 (±3.86) with no vibrator activated (21 measures), which is not statistically significant (t = −0.64, P = 0.52). These results show that the vibrators are fMRI compatible when placed 20 cm or more away from the head, which is the case in most experiments since the vibrators are likely to be strapped on the legs or arms.

Concerning the frequency calibration, the mean vibration frequency as a function of the percentage of aperture is given in Figure 4b for the calibration both inside and outside the fMRI machine.

The results of the calibration measures showed that our pneumatic device could generate vibrations from 40 to 100/120 Hz (100 Hz inside and 120 Hz outside the scanner), which is the appropriate frequency range for activating muscle spindles. Figure 4b shows that for aperture values from 40 to 70%, the vibration frequency was approximately a linear function of the aperture of the flowmetric gate (r 2 = 0.93). The vibration frequency reached an asymptote at more than 70% aperture. This was true for the vibrator inside as well as outside the scanner. The effects of aperture and environment (inside versus outside the scanner) were both statistically significant (F(1,11) = 59.79, P < 0.001, and F(1,11) = 46.9, P < 0.001, respectively). The vibration frequency was slightly lower inside than outside the scanner, independently of the percentage of aperture of the flowmetric gate. This difference was more pronounced for vibration frequencies higher than 90 Hz than for vibration frequencies between 40 and 90 Hz (on average, 17 and 7.6 Hz, respectively). However, there was no significant interaction between the percentage of aperture and the environment (inside/outside the scanner) of the vibrator (F(1,11) = 2.62, P = 0.14). This effect of the environment of the vibrator is probably due to the off‐centered mass. Indeed, barite is not completely amagnetic, although its magnetic susceptibility value is close to zero. It is slightly diamagnetic. This property is most likely responsible for the slow‐down of the off‐centered mass inside the scanner. Despite this drawback, the vibration frequency inside the scanner is still in the efficient range for inducing reliable illusions of movement (i.e., 30–100 Hz), and the vibrator does not impair the fMRI signal in any noticeable way.

Kinesthetic Test

Figure 5 gives the velocity of the illusory movement as a function of the vibration frequency of the pneumatic vibrator.

Figure 5.

Figure 5

Open in a new tab

Mean velocity (and standard errors) of illusory movements induced by a pneumatic vibrator as a function of vibration frequency (data from six participants).

The results show that the pneumatic vibrator was effective in inducing movement illusions. The velocity of these illusions increased with vibration frequency reaching its maximum around 90 Hz. At 100 Hz, the speed of the induced illusion started to decrease. We did not test higher frequencies for two reasons. First, our apparatus cannot go beyond 120 Hz (see Fig. 4b) and second there is no reason to go any further since reliable movement illusions of various speeds can be obtained with vibration frequencies in the 40–100 Hz range. Previous studies using electromagnetic vibrators have obtained very similar results [see Roll and Vedel, 1982; Vedel and Roll, 1983]. The present results show that our pneumatic device is able to replicate previously kinesthetic data acquired with classical vibrators.

DISCUSSION AND CONCLUSION

The present article presents a new pneumatic vibration device that is compatible with the high strength magnetic field environment of fMRI machines. We show that this noninvasive device does not interfere with the magnetic field and does not distort the images. Moreover, this pneumatic device operates as intended in the fMRI environment; that is, it can produce vibrations between 30 and 100 Hz, inside or outside the scanner, and its efficiency is not significantly affected by the magnetic field. We also show that this pneumatic device can reliably stimulate muscle proprioception and generate reliable illusions of movements that are comparable to those generated by classical electromagnetic vibrators [Roll and Vedel, 1982; Vedel and Roll, 1983].

In the specific apparatus we have designed, six vibrators can be used simultaneously and controlled independently. Given the size of our prototype (5 × 6 cm), a maximum of four vibrators can be attached to small articulations like the wrist, which is sufficient for generating complex illusions of movements [e.g., drawing letters, digits, or geometrical figures, see Albert et al., 2006; Roll and Gilhodes, 1995]. The vibrators can also be distributed over different joints of the same limb or different body parts.

A number of alternative fMRI‐compatible devices have been proposed for the study of somesthesia [e.g., Briggs et al., 2004; Gelnar et al., 1998; Gizewski et al., 2005; Graham et al., 2001; Harrington et al., 2000; Stippich et al., 1999]. Until now, however, none of them have been specifically designed to produce constant, replicable, and efficient muscle proprioceptive stimulation simultaneously on different body areas. Piezoelectric devices typically do not generate adequate vibration amplitude for robust muscle spindle stimulation to occur. When they do, they cannot stimulate more than one body region at a time. MVD cannot produce replicable stimulations within the fMRI machine since the parameters of the stimulation vary with the position and orientation of the stimulators in the magnetic field. Finally, the pneumatic stimulators proposed so far produce vibrations that are not in the optimal frequency range (40–100 Hz) or duration range for generating illusions of movement. Of course, these devices are likely to be adapted and scaled up to circumvent these problems in the future. For now, our pneumatic device provides a promising alternative that was successfully used in three recent fMRI studies [Duclos et al., 2007; Kavounoudias et al., 2008; Romaiguère et al., 2003] where it induced an illusion of hand movement and generated significant activations in the sensorimotor and motor‐related areas. Note that this apparatus is also compatible with magnetoencephalography (MEG), as shown in a recent study [Casini et al., 2006]. One of the differences between fMRI and MEG is that MEG machines are not as loud as fMRI machines. Even with silencers, our pneumatic vibrators can be heard by the participants during MEG experiments. To avoid interferences from noise, we replaced the silencers by polyethylene tubes to evacuate the air outside the scanning room. The participants were also equipped with earplugs and headphones producing continuous white noise.

In the present article, we tested our pneumatic vibrator on a simple kinesthetic illusion but this device can also be used to study illusory movements that are more complex. For example, several vibrators can be used in parallel (up to six vibrators with our apparatus) on the same body segment, each set to a specific frequency/time range. Such a configuration allows one to mimic the stretching of different muscle groups during complex movements like drawing [Albert et al., 2006; Roll and Gilhodes, 1995]. For example, Roll and Gilhodes [ 1995] were able to generate the illusion of drawing geometrical forms or of cursive writing of letters and numbers by activating four groups of muscle tendons of the right wrist. Each letter or digit was coded by a specific vibration sequence of the four vibrators. The parameters of the different patterns of vibration (frequency variation, onset, and duration) were determined by a mechanical model along with previous experimental data on illusory movements [Gilhodes et al., 1993]. Albert et al., [ 2006] did a similar experiment using patterns of vibration that were directly inspired from Ia fiber activity during imposed “writing‐like” movements. In both studies, the participants were able to recognize and name the symbols evoked by the vibration, which suggests that proprioception, like the other sensory channels, may contribute to high‐level cognition. This assumption could be easily tested in an fMRI study using our pneumatic device.

Our pneumatic device can also be used to investigate other kinesthetic phenomena, such as the antagonist vibratory response [e.g., Calvin‐Figuière et al., 2000] or the motor posteffect that develops after the release of a sustained muscle contraction [e.g., Duclos et al., 2004]. This device can also be useful for brain imaging studies targeting multisensory integration [e.g., tactile/proprioceptive cooperation, Kavounoudias et al., 2001; Kavounoudias et al., 2008] or functional recovery following kinesthetic rehabilitation [Neiger et al., 1983; Roll, 1998].

Acknowledgements

We thank Johannes Ziegler for helpful comments, Jean‐Luc Anton, Bruno Nazarian, Muriel Roth and Nadia Tir, for their technical contribution.

REFERENCES

  1. Albert F,Bergenheim M,Ribot‐Ciscar E,Roll JP ( 2006): The Ia afferent feedback of a given movement evokes the illusion of the same movement when returned to the subject via muscle tendon vibration. Exp Brain Res 172: 163–174. [DOI] [PubMed] [Google Scholar]
  2. Briggs RW,Dy‐Liacco I,Malcom MP,Lee H,Peck KK,Gopinath KS,Himes NC,Soltysik DA,Browne P,Tran‐Son‐Tay R ( 2004): A pneumatic vibrotactile stimulation device for fMRI. Magn Reson Med 51: 640–643. [DOI] [PubMed] [Google Scholar]
  3. Burke D,Hagbarth KE,Lofstedt L,Wallin BG ( 1976a): The responses of human muscle spindle endings to vibration of non‐contracting muscles. J Physiol 261: 673–693. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Burke D,Hagbarth KE,Lofstedt L,Wallin BG ( 1976b): The response of human muscle spindle endings to vibration during isometric contraction. J Physiol 261: 695–711. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Calvin‐Figuière S,Romaiguère P,Roll JP ( 2000): Relations between the directions of vibration‐induced kinesthetic illusions and the pattern of activation of antagonist muscles. Brain Res 881: 128–138. [DOI] [PubMed] [Google Scholar]
  6. Casini L,Romaiguère P,Ducorps A,Schwartz D,Anton JL,Roll JP ( 2006): Cortical correlates of illusory hand movement perception in humans: A MEG study. Brain Res 1121: 200–206. [DOI] [PubMed] [Google Scholar]
  7. Courbière M,Valette PJ,Pilleul F ( 2005): X‐ray investigation of the small bowel in 2005. Acta Endoscopica 35: 285–301. [Google Scholar]
  8. Duclos C,Roll R,Kavounoudias A,Roll JP ( 2004): Long lasting body learning following neck muscles isometric contractions. Exp Brain Res 158: 58–66. [DOI] [PubMed] [Google Scholar]
  9. Duclos C,Roll R,Kavounoudias A,Roll JP ( 2007): Cerebral correlates of the “Kohnstamm phenomenon”: An fMRI study. Neuroimage 34: 774–783. [DOI] [PubMed] [Google Scholar]
  10. Eklund G ( 1972): Position sense and the state of contraction: The effects of vibration. J Neurol Neurosurg Psychiatry 35: 606–611. [Google Scholar]
  11. Gelnar PA,Krauss BR,Szeverenyi NM,Apkarian AV ( 1998): Fingertip representation in the human somatosensory cortex: An fMRI study. Neuroimage 7: 261–283. [DOI] [PubMed] [Google Scholar]
  12. Gilhodes JC,Roll JP,Tardy‐Gervet MF ( 1986): Perceptual and motor effects of agonist‐antagonist muscle vibration in man. Exp Brain Res 61: 395–302. [DOI] [PubMed] [Google Scholar]
  13. Gilhodes JC,Coiton Y,Roll JP,Ans B ( 1993): Proprio‐muscle coding of kinesthetic sensation. Experimental approach and mathematical modelling. Biol Cybern 68: 509–519. [DOI] [PubMed] [Google Scholar]
  14. Gizewski ER,Koeze O,Uffmann K,de Greiff A,Ladd ME,Forsting M ( 2005): Cerebral activation using MR‐compatible piezoelectric actuator with adjustable vibration frequencies and in vivo wave propagation control. Neuroimage 24: 723–730. [DOI] [PubMed] [Google Scholar]
  15. Goodwin GM,McCloskey DI,Matthews PCB ( 1972): The contribution of muscle afferents to kinesthesia shown by vibration‐induced illusions of movement and by the effect of paralyzing joint afferents. Brain 95: 705–748. [DOI] [PubMed] [Google Scholar]
  16. Graham SJ,Staines WR,Nelson A,Plewes DB,McIlroy WE ( 2001): New devices to deliver somatosensory stimuli during functional MRI. Magn Reson Med 46: 436–442. [DOI] [PubMed] [Google Scholar]
  17. Harrington GS,Wright CT,Hunter Downs J III ( 2000). A new vibrotactile stimulator for functional MRI. Hum Brain Mapp 10: 140–145. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Jones AL ( 1988). Motor illusions: What do they reveal about proprioception? Psychol Bull 103: 72–86. [DOI] [PubMed] [Google Scholar]
  19. Kavounoudias A,Roll R,Roll JP ( 2001): Foot sole and ankle muscle inputs contributes jointly to human erect posture regulation. J Physiol 532: 869–978. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kavounoudias A,Roll JP,Anton JL,Nazarian B,Roth M,Roll R ( 2008): Proprio‐tactile integration for kinesthetic perception: An fMRI study. Neuropsychologia 46: 567–575. [DOI] [PubMed] [Google Scholar]
  21. Li KCP,Tart RP,Fitzsimmons JR,Storm BL,Mao J,Rolfes RJ ( 1991): Barium sulfate suspension as a negative oral MRI contrast agent: In vitro and human optimization studies. Magn Reson Imaging 9: 141–150. [DOI] [PubMed] [Google Scholar]
  22. Marsden CD ( 1982): The mysterious functions of the basal ganglia: The Robert Wartenberg Lecture. Neurology 32: 514–539. [DOI] [PubMed] [Google Scholar]
  23. Matthews PBC ( 1982): Mammalian Muscle Receptors and Their Central Actions. London: E. Arnold Publishers Ltd. [Google Scholar]
  24. Neiger H,Gilhodes JC,Roll JP ( 1983): Méthode de rééducation motrice par assistance proprioceptive vibratoire. Partie II, Annales de Kinésithérapie 10: 16–28. [Google Scholar]
  25. Roll JP ( 1998): Les muscles, organes de la perception. Pour La Sci 248: 92–99. [Google Scholar]
  26. Roll JP,Vedel JP ( 1982): Kinesthetic role of muscle afferents in man, studies by tendon vibration and microneurography. Exp Brain Res 47: 177–190. [DOI] [PubMed] [Google Scholar]
  27. Roll JP,Gilhodes JC ( 1995): Proprioceptive sensory codes mediating movement trajectory perception: Human hand vibration‐induced drawing illusions. Can J Physiol Pharmacol 73: 295–304. [DOI] [PubMed] [Google Scholar]
  28. Roll JP,Vedel JP,Ribot E ( 1989): Alteration of proprioceptive messages induced by tendon vibration in man: A microneurographic study. Exp Brain Res 76: 213–222. [DOI] [PubMed] [Google Scholar]
  29. Roll R,Velay JL,Roll JP ( 1991): Eye and neck proprioceptive messages contribute to spatial coding of retinal input in visually oriented activities. Exp Brain Res 85: 423–431. [DOI] [PubMed] [Google Scholar]
  30. Roll JP,Gilhodes JC,Roll R,Harlay F ( 1996). Are proprioceptive sensory inputs combined into a “Gestalt”? In: Inui T,McClelland JL, editors. Attention and Performance, Information Integration in Perception and Communication. Cambridge: MIT Press; 291–314. [Google Scholar]
  31. Roll JP,Albert F,Ribot‐Ciscar E.Bergenheim M ( 2004): “Proprioceptive signature” of cursive writing in humans: A multi‐population coding. Exp Brain Res 57: 359–368. [DOI] [PubMed] [Google Scholar]
  32. Romaiguère P,Anton JL,Roth M,Casini L,Roll JP ( 2003): Motor and parietal cortical areas both underlie kinaesthesia. Cogn Brain Res 16: 74–82. [DOI] [PubMed] [Google Scholar]
  33. Stippich C,Hofmann R,Kapfer D,Hempel E,Heiland S,Jansen O,Sartor K ( 1999): Somatotopic mapping of the human primary somatosensory cortex by fully automated tactile stimulation using functional magnetic resonance imaging. Neurosci Lett 277: 25–28. [DOI] [PubMed] [Google Scholar]
  34. Schenck J ( 1996): The role of magnetic susceptibility in magnetic resonance imaging: MRI magnetic compatibility of the first and second kinds. Med Phys 23: 815–850. [DOI] [PubMed] [Google Scholar]
  35. Teasdale N,Forget R,Bard C,Paillard J,Fleury M,Lamarre Y ( 1993): The role of proprioceptive information for the production of isometric forces and for handwriting tasks. Acta Psychologica 82: 179–191. [DOI] [PubMed] [Google Scholar]
  36. Vedel JP,Roll JP ( 1983): Muscle spindle contribution to the coding of motor activities in man. Exp Brain Res Suppl 7: 253–265. [Google Scholar]

Articles from Human Brain Mapping are provided here courtesy of Wiley

RESOURCES