- Email: [email protected]
Inhibition of the anterior intraparietal area and the dorsal premotor cortex interfere with arbitrary visuo-motor mapping
Clinical Neurophysiology 121 (2010) 408–413
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Inhibition of the anterior intraparietal area and the dorsal premotor cortex interfere with arbitrary visuo-motor mapping Marco Taubert a,b,d, Manuel Dafotakis a, Roland Sparing c, Simon Eickhoff a, Siegfried Leuchte d, Gereon R. Fink a,c, Dennis A. Nowak a,c,e,* a
Institute of Neuroscience and Medicine (INB3), Cognitive Neurology Section, Research Centre Jülich, Jülich, Germany Max Planck Institute for Human Cognitive and Brain Sciences, Department of Cognitive Neurology, Leipzig, Germany Department of Neurology, University of Cologne, Cologne, Germany d Department of Sport Science, Martin-Luther-University Halle-Wittenberg, Halle, Germany e Neurologische Fachklinik Kipfenberg, Kipfenberg, Germany b c
a r t i c l e
i n f o
Article history: Accepted 8 November 2009 Available online 9 December 2009 Keywords: Transcranial magnetic stimulation Dorsal premotor cortex Anterior intraparietal sulcus
a b s t r a c t Objective: The contribution of the human anterior intraparietal area and the dorsal premotor cortex to arbitrary visuo-motor mapping during grasping were tested. Methods: Trained right-handed subjects reached for and pincer-grasped a cube with the right hand in the absence of visual feedback after the cube location had been displayed for 200 ms. During the reaching movements, the colour of the cube changed and visual feedback about the change of colour was provided for 100 ms at 500 ms after movement onset (at the time of peak grasp aperture). Depending on colour, subjects were instructed to either pincer-grasp the cube in a horizontal or vertical grasp position with the latter necessitating wrist rotation (experiment 1) or to pincer-grasp and transport the cube to either a left or right target position (experiment 2). Within two consecutive 200 ms time windows (TMS 1 and 2) starting 500 ms and 700 ms after movement onset, respectively, double pulses of supra-threshold transcranial magnetic stimulation (inter-stimulus interval: 100 ms) were delivered over (i) the left primary motor cortex (90° vertically angulated coil position, control stimulation), (ii) the left dorsal premotor cortex or (ii) the left anterior intraparietal area. Results: Compared to control stimulation, stimulation of the anterior intraparietal area, but not of the dorsal premotor cortex, at TMS 1 delayed the times to wrist rotation (experiment 1) and hand transport (experiment 2). Compared to control stimulation, stimulation of the dorsal premotor cortex, but not of the anterior intraparietal area, at TMS 2 delayed both wrist rotation (experiment 1) and hand transport (experiment 2). Conclusions: We contend that the anterior intraparietal area and the dorsal premotor cortex are both involved albeit at different phases during the mapping of arbitrary visual cues with goal directed grasp and transport movements. Significance: These data add to the current understanding of how human cortical areas work in concert during manual activities. Ó 2009 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction Reaching for and grasping an object is part of our daily motor repertoire and is usually generated effortlessly despite a huge variety of potential sources of error. Errors may result from sudden
* Corresponding author. Address: Klinik Kipfenberg, Neurochirurgische und Neurologische Fachklinik, Kindinger Strasse 13, D-85110 Kipfenberg, Germany. Tel.: +49 (0) 8465 175 100; fax: +49 (0) 8465 175 184. E-mail addresses: [email protected], dennis.nowak@ uk-koeln.de (D.A. Nowak).
contextual changes in the environment or from intrinsic noise within the sensory-motor system causing inaccurate movement execution. The accuracy of goal-directed motor performance depends on rapid detection and correction of these errors necessitating online comparison between the issued motor command, current sensory feedback and the specific aim of the intended action. One key issue underlying goal-directed interactions with the environment is the ability to form rapid associations between environmental cues and intended actions. The rapid matching of environmental cues with purposeful actions is referred to as arbitrary visuo-motor mapping (Wise and Murray, 2000).
1388-2457/$36.00 Ó 2009 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.clinph.2009.11.011
M. Taubert et al. / Clinical Neurophysiology 121 (2010) 408–413
Human reaching and grasping seem to involve a widespread parieto-frontal cortical motor network including the anterior intraparietal sulcus (aIPS, putative homologue of macaque anterior intraparietal area), the dorsal premotor cortex (PMd), the ventral premotor cortex and the primary motor cortex (for a recent review see Castiello, 2005). To date, however, less is known about the neural correlates involved in arbitrary visuo-motor mapping. The present study focused on the contributions of the aIPS and PMd to arbitrary visuo-motor mapping during grasping movements. Electrophysiological recordings in the macaque indicate that neurons within the anterior intraparietal area are recruited for visual and/or tactile object discrimination and are preferentially activated for various hand configurations when grasping objects of different shape (Grefkes and Fink, 2005; Murata et al., 2000). Humans with lesions of the aIPS show difficulties to shape their hand according to the intrinsic object characteristics during visually guided reach-to-grasp movements (Binkofski et al., 1998). A virtual lesion of the left (dominant) aIPS in healthy right-handed subjects, induced by transcranial magnetic stimulation (TMS), disrupts the online correction of right hand grasp formation during visually guided reach-to-grasp movements toward an object that unexpectedly changes its size (Tunik et al., 2005; Rice et al., 2006). Thus, aIPS seems to be involved in at least two different processes: (i) the transformation of intrinsic object properties, e.g., size and shape, into a motor plan, and (ii) the online detection of a mismatch between the intended action and environmental context that is necessary to correct grasp formation during visually guided grasping. Lesions of the PMd may result in impairments to reproduce previously acquired visuo-motor mappings or to learn novel mappings, such as linking an arbitrary colour cue to a grasping task, both in macaques and humans (Petrides, 1982, 1985; Nixon et al., 2004). TMS studies have shown that the left (dominant) PMd is essential for the predictive scaling of grasping forces based on learned associations between arbitrary colour cues and the mass of an object to be lifted with either the right or left hand in right-handed subjects (Chouinard et al., 2005; Nowak et al., 2009). Together, these studies indicate that PMd is involved in the context-specific selection of appropriate motor commands based on arbitrary visuo-motor mappings with external cues during grasping. The specific role of the aIPS for (i) the integration of the relevant object features into central motor control programs and (ii) the dynamic online detection of changes in intrinsic object characteristics during grasping (Tunik et al., 2005) make an involvement of aIPS in arbitrary visuo-motor mapping very likely. One putative contribution of aIPS is that it is involved in the online detection and/or integration of arbitrary visual stimuli to be matched with particular actions based on learned associations as provided by PMd. Within this context, aIPS and PMd may be differentially involved at consecutive time intervals in the process of arbitrary visuo-motor mapping during grasping. To test this hypothesis, trained right-handed subjects reached for and pincer-grasped a cube in the absence of visual feedback after the location of the cube had been displayed for 200 ms. During the reach movement the colour of the cube changed and visual feedback about the current cube colour was provided for 100 ms at the time of peak grasp aperture. Depending on the colour, subjects either had to grasp the cube in a horizontal or vertical pincer-grasp position with the latter necessitating hand rotation (experiment 1) or to grasp and transport the cube to either a left or right target position (experiment 2). Within either 0–200 ms (TMS 1) or 200–400 ms (TMS 2) after peak grasp aperture TMS double pulses (inter-stimulus interval: 100 ms) were delivered over (i) the left primary motor cortex (M1, 90° vertically angulated coil position, control stimulation), (ii) the left PMd or (iii) the left aIPS. We hypothesized that, compared to control stimulation over M1, TMS over aIPS and PMd interfered with
409
arbitrary visuo-motor mapping at different time intervals, respectively. 2. Methods 2.1. Participants Ten right-handed (Oldfield, 1971) male subjects (mean age ± SD: 30 ± 7 years) with normal or corrected to normal vision participated. Seven subjects participated in experiment 1; eight subjects participated in experiment 2. Five subjects participated in both experiments. All subjects were naïve to the specific purpose of the study and provided written informed consent. Subjects were screened for neurological disorders and contraindications for TMS. Each subject was able to detect and discriminate the different colours (red, white and black) of the cubes used in experiments 1 and 2. The experimental procedures were approved by the local Ethics committee. 2.2. Neuronavigation procedure A T1-weighted, high-resolution three-dimensional volumetric magnetic resonance imaging (MRI) scan of each subject’s brain was acquired (3 T, Trio, Siemens, Erlangen, Germany; 176 contiguous 1 mm thick sagittal slices). The subject’s head and the individual MRI scan were carefully co-registered in a common reference frame with an infrared optical-tracking neuronavigational system (Polaris System, Northern Digital, Waterloo, Ontario, Canada, and Brainsight software, Rogue Research, Montreal, Canada). The anatomical landmarks used were the tip and the alar wings of the nose, the nibs of the tragus of both ears, and the internal angles of both eyes. The following procedure was used to locate the stimulation site and to monitor the position of the TMS coil over the left M1, left PMd and left aIPS, respectively: First, subject’s structural MRI was transferred into the Montreal Neurological Institute (MNI) coordinate system. Second, probabilistic locations for M1 (x = 31, y = 22, z = 52), PMd (x = 24, y = 3, z = 62) and aIPS (x = 42, y = 39, z = 44) of the left hemisphere were derived based on previous studies (Binkofski et al., 1999; Chouinard et al., 2005; Grefkes and Fink, 2005). Third, the probabilistic locations were transferred to the subject’s brain coordinate space and marked on the three-dimensional MRI scan for stimulation purposes. Control stimulation was performed by positioning the TMS coil in a 90° angulated orientation over the hand area of M1. Single pulse TMS was used to verify correct coil placement with electromyographic activity recorded from the contralateral first dorsal interosseus muscle (see below). The neuronavigational system was then used to navigate the TMS coil over the probabilistic locations marked within the subject’s MRI. The TMS coil was fixed in position over the scalp by a custom made positioning system throughout each session. During the experiments, the centre of the coil was continuously monitored to be over the site of interest and minor deviations were corrected prior to each experimental block. A maximum deviation of 3 mm was tolerated. For stimulation of the PMd and aIPS, the TMS coil was positioned tangential to the surface of the skull with the short axis of the figure-of-eight coil angled 45° relative to the interhemispheric fissure and approximately perpendicular to the central sulcus with the handle pointing lateral and caudal. For control stimulation the TMS coil was placed in 90° vertically angulated coil position at the skull over left M1. That is the magnetic coil was placed vertically on the scalp so that the lateral aspect of the coil touched the scalp and the magnetic field ran tangentially to the scalp. This positioning did not interfere with neural processing within M1. Head movements were restricted by the use of a chin rest and two additional positioning frames.
410
M. Taubert et al. / Clinical Neurophysiology 121 (2010) 408–413
2.3. Transcranial magnetic stimulation (TMS) TMS was performed using a 70-mm figure-of-eight coil and two Magstim 200 connected via a BiStim Module (Magstim Company, Dyfed, UK). The following procedure was performed to obtain the resting motor threshold for each subject. The coil was placed tangentially over the left M1 at the optimal site to elicit a response of the right first dorsal interosseus muscle. The optimal site was defined as the location where stimulation at a slightly supra-threshold intensity elicited the largest motor evoked potential. Electromyographic activity was recorded using silver–silver–chloride electrodes positioned in a belly-tendon technique on the skin overlying the right first dorsal interosseus muscle. The electromyographic signal was amplified, filtered (50–2000 Hz) and digitized at a sampling rate of 5000 Hz. The resting motor threshold was defined for each participant as the lowest stimulator output intensity that elicited motor evoked potentials with peak-to-peak amplitude of at least 50 lV in at least 5 of 10 trials. Two single pulses of TMS (inter-stimulus interval of 100 ms) were applied at an intensity of 100% of the resting motor threshold over M1 and PMd, but 120% of the resting motor threshold over aIPS as the latter is located in the intraparietal sulcus and greater stimulation intensities are necessary to reach the aIPS (Stokes et al., 2005). The application of TMS pulses at intervals of 100 ms induces long interval cortical inhibition (LICI), which is induced by an enhancement of GABA-B receptor activity (Werhahn et al., 1999). 2.4. Kinematic motion analysis Reach-to-grasp movements were recorded using an ultrasonic motion measurement system (CMS 20S, Zebris, Isny, Germany). The system uses three microphones to continuously calculate the three-dimensional spatial positions of small ultrasound emitting markers (diameter: 5 mm) attached via flexible cables to the moving segments of the upper limb. Three position markers were fixed to the styloid process of the radius, the dorsal tip of the index finger and the dorsal tip of the thumb. The position markers placed on the dorsal tips of the index finger and thumb allowed for the registration of the opening and closure of the pincer-grasp. Spatial coordinates of each position marker were sampled at a frequency of 100 Hz and a spatial resolution of 0.1 mm. 2.5. Experimental procedure Subjects were seated in front of a table with their chin positioned in a chin rest to avoid head movements. Subjects wore headphones providing white noise thereby eliminating auditory information during task performance. At the beginning of each trial subjects placed the wrist of their right hand on a start button and pincer-grasped a piece of tape between thumb and index finger to standardize initial hand position. Subjects were instructed to reach for and pincer-grasp a cube positioned on a turning table 42 cm away from the start button and 20 cm below shoulder level between the index finger and thumb. Two cubes of equal dimensions (2 2 2 cm) were fixed on a turning table mounted on the shaft of a motor (model TMCM; Trinamic, Hamburg, Germany) separated by a vertical wall in the middle of the table. The cubes had different shapes in experiments 1 and 2. In experiment 1, subjects had to grasp the cube between the index finger and thumb in a horizontal or vertical grasp (Fig. 1). During a vertical grasp the index finger should be placed on the top plane and the thumb on the bottom plane of the cube. Thus, the cube in experiment 1 was positioned in 30° angulation on the turning table (see Fig. 1). In experiment 2, subjects always grasped the cube in a horizontal grasp. Thus, the cube was positioned on its bottom plane on the turning table in experiment 2. In both experiments, two similar shaped,
but different-coloured, cubes were placed on each side of the wall so that subjects were able to see the cube on the front side of the turning table. The motor of the turning table was programmed to turn the table for 180° after movement initiation (release of the start button) to present the cube on the other side of the wall (time for motor to complete rotation: 100 ms). Visual feedback was controlled by liquid crystal shutter glasses (Plato System; Translucent Technologies, Toronto, Ontario, Canada). The shutter glasses (time to switch between light scattering opaque and transparent state: 4 ms) remained opaque during each trial except for a 200 ms window prior to movement onset, which served as a go-signal to initiate the movement and a 100 ms window initiated 500 ms after movement onset (corresponding to the time of peak grasp aperture) (see Fig. 1). Rotation of the turning table presented the cube on the back side of the wall, which was visible within the second 100 ms window the shutter glasses were transparent. The shutter glasses then turned opaque again for the remainder of the trial. In experiment 1 (n = 7), subjects saw a red cube (on the front side of the turning table) during the first time window and a different-coloured, white or black, cube (on the back side of the turning table) during the second time window the shutter glasses turned transparent 500 ms after movement onset. The colour of the cube indicated that subjects had to pincer-grasp the cube in a horizontal grasp along its lateral surfaces (black cube) or vertical grasp along its upper and lower surfaces (white cube). Grasping the cube in a vertical pincer-grasp required a rapid wrist rotation before object contact. In experiment 2 (n = 8), similar to the first experiment, subjects saw a red cube (on the front side of the turning table) in the first time window and different-coloured, white or black, cube (on the back side of the turning table) in the second time window. Here, the colour of the cube indicated that the cube had to be grasped in a horizontal pincer-grasp at its lateral surfaces and rapidly transported and placed into one of two boxes on the left (white cube) or right side (black cube) of the turning desk (distance 15 cm). Note that subjects had to finish the reach-to-grasp movement before transporting the cube to either of the two boxes. During the reach-to-grasp movement in both experiments, double pulse TMS (inter-stimulus interval of 100 ms; Rice et al., 2006) were applied (i) 500–700 ms (TMS 1) and (ii) 700–900 ms (TMS 2) after movement onset. Note that TMS 1 included the period the shutter glasses turned transparent for a second time (see Fig. 1). TMS 1 during the transparent phase of the shutter glasses was thought to interfere with the detection and integration of the arbitrary colour cue. TMS 2 was thought to interfere with the selection of motor commands for wrist rotation (experiment 1) or transport movement (experiment 2) based on arbitrary colour cues. TMS was applied over left M1, PMd and aIPS on separate days. The sequence of TMS application was randomly assigned to each subject. On each day subjects performed 5 blocks of 20 trials. Within each block an equal number of black and white cubes were presented for TMS 1 and TMS 2 in random order. This procedure ensured that subjects were unable to predict cube colour of the subsequent trial. Prior to each experimental session, subjects were trained by performing practice session (4 blocks of 20 trials each) to learn the task requirements. 2.6. Data analysis The X-, Y- and Z-directions of the position marker coordinates refer to the medio-lateral, antero-posterior and vertical directions with regard to the subject performing the task. The software (3DA, MedCom, Munich, Germany) also allowed for the calculation of distances between single position markers. For each reach-tograsp movement, we determined movement onset as the first detectable 5% change of wrist velocity in the Z-direction from base-
M. Taubert et al. / Clinical Neurophysiology 121 (2010) 408–413
411
Fig. 1. Illustration of the experimental paradigms for experiments 1 and 2. Subjects were seated in front of a table. At the beginning of each trial subjects placed the wrist of their right hand on a start button and pincer-grasped a piece of tape between thumb and index finger to standardize initial hand position. Subjects wore headphones which provided white noise during task performance. Subjects were instructed to reach for and pincer-grasp a cube positioned on a turning table between the index finger and thumb. Two cubes of equal dimensions were fixed on a turning table separated by a vertical wall in the middle of the table. The cubes were placed on each side of the wall so that subjects were able to see the cube on the front side of the turning table. The table turned 180° after movement initiation (release of the start button) to present the cube on the other side of the wall. Visual feedback was controlled by liquid crystal shutter glasses. The shutter glasses remained opaque during the trial except for a 200 ms window prior to movement onset, which served as a go-signal to initiate the movement and a 100 ms window initiated 500 ms after movement onset. The colour of the cube displayed at the time of peak grasp aperture indicated to either grasp the cube in a horizontal or vertical position (experiment 1) or grasp and transport the cube to a left or right target position (experiment 2).
line obtained from the marker fixed to the styloid process of the radius. Performance in experiment 1 was quantified by assessing the time (in ms) from the onset of the second shutter glasses opening (500 ms after movement onset) to the onset of wrist rotation. Only the trials in which the wrist was required to rotate from a horizontal to a vertical grasp position were included in the data analysis. The onset of wrist rotation was determined as the first detectable 5% change in the angular velocity of the index finger marker within the X-plane. Performance in experiment 2 was quantified by assessing (i) the time needed (in ms) from the onset of the second shutter glasses opening to object contact (minimal distance between thumb and index finger markers and velocity offset) and (ii) the time (in ms) from object contact to the onset of the transport movement to the left or right box (first detectable 5% change of index finger velocity in the X-direction from baseline). All parameters were averaged across all trials performed by each participant. Mean values were calculated for each subject within both experiments. Repeated measures analyses of variance (ANOVA) with a level of significance of P < 0.05 were calculated to assess the effect of ‘‘stimulation site” (TMS left M1, left PMd and left aIPS) and ‘‘time” (TMS1 and TMS2) on reaction times. Post-hoc t-tests were corrected for multiple testing according to Bonferroni. 3. Results 3.1. Experiment 1 Fig. 2a illustrates average times (+1 SEM) from the onset of visual feedback about cube colour (onset of the second shutter glasses opening, see Fig. 1) to the initiation of wrist rotation. It is evident that when applied over the left aIPS, but not over the left PMd, TMS 1 delayed the times needed to perform the wrist rotation compared to control stimulation over the left M1 (with 90°
angulated coil orientation). In contrast, TMS 2 prolonged the times needed to rotate the hand when applied over the left PMd, but not over the left aIPS, compared to control stimulation. These effects were significant: repeated measures ANOVA revealed no significant effects of the single factors ‘‘stimulation site” or ‘‘time”. However, there was a significant effect of the interaction ‘‘stimulation site” ‘‘time” on times to hand rotation (F1,6 = 14.6; P < 0.001). Post-hoc t-tests showed that for TMS 1 the times needed to rotate the wrist were longer after TMS over the left aIPS compared to stimulation over the left PMd or control stimulation over left M1 (P < 0.01 for each comparison). In addition, for TMS 2 the times needed to rotate the wrist were longer after TMS over the left PMd compared to TMS over the left aIPS or the left M1 (P < 0.01 for each comparison). 3.2. Experiment 2 In experiment 2, the times from the onset of the second shutter glasses opening to object contact were not significantly prolonged by TMS, regardless of the site or time of stimulation. Repeated measures ANOVA detected neither a significant effect of the single factors ‘‘stimulation site” or ‘‘time” nor their interactions on times to object contact. Fig. 2b illustrates average times (+1 SEM) from object contact to the onset of the transport movement of the hand to the left or right box. Fig. 2b shows average times (+1 SEM) from the onset of the second shutter glasses opening to the time of hand transport initiation. Repeated measures ANOVA detected no significant effects of the single factors ‘‘stimulation site” or ‘‘time” on times to hand transport. A significant interaction ‘‘stimulation site” ‘‘time” on times to hand transport (F1,7 = 14.6; P < 0.001) suggested differential effects of stimulation site on times to hand transport depending on the time of stimulation. Pot hoc tests revealed that for TMS 1
412
M. Taubert et al. / Clinical Neurophysiology 121 (2010) 408–413
about the actual colour of the object was displayed for 100 ms at the time of peak grasp aperture. The colour indicated that subjects either had to pincer-grasp the cube in a horizontal or vertical grasp position with the latter necessitating a wrist rotation (experiment 1) or to pincer-grasp and transport the cube to either a left or right target position (experiment 2). TMS was used to modulate neural processing of (i) the left primary motor cortex (90° vertically angulated coil position, control stimulation), (ii) the left PMd or (iii) the left aIPS during (TMS 1) or after (TMS 2) the presentation of the visual colour cue. These differential time points for TMS were introduced to detect potential sequential processing in aIPS and PMd. Stimulation of left aIPS, but not of left PMd, disturbed the times to the onset of wrist rotation (experiment 1) and onset of hand transport (experiment 2) when given at TMS 1. In contrast, stimulation of left PMd, but not left aIPS, at TMS 2 delayed the times to the initiation of hand rotation (experiment 1) and hand transport (experiment 2). These data add new evidence to the functional temporal coupling between aIPS and PMd during visuo-motor mapping.
4.1. The role of the aIPS for arbitrary visuo-motor mapping
Fig. 2. (a) Average times (+1 SEM) from the onset of the second time window of shutter glasses opening to the initiation of wrist rotation (experiment 1, N = 7) and (b) average times from object contact to the initiation of the wrist transport movement (experiment 2, N = 8). The times to the initiation of wrist rotation (experiment 1) and the times to the initiation of wrist transport (experiment 2) were significantly longer following early TMS 1 over left aIPS compared to TMS over left PMd or control stimulation over left M1. In contrast, the times to the initiation of wrist rotation (experiment 1) and the times to the initiation of wrist transport (experiment 2) were significantly longer following late TMS 2 over left PMd compared to TMS over left aIPS or control stimulation over left M1. For the definition of the time points of TMS application within the time course of the experiment refer to Fig. 1. Significant differences are indicated (**P < 0.01).
times from the second shutter glasses opening to initiation of hand transport were longer when TMS was applied over the left aIPS, compared to stimulation over the left PMd or control stimulation over left M1 (P < 0.01 for each comparison). In contrast, for TMS 2 times to hand transport onset were longer when TMS was applied over the left PMd, compared to stimulation over the left aIPS or control stimulation over the left M1 (P < 0.01 for each comparison). 4. Discussion The present experiments were designed to investigate the contribution of the (dominant) left aIPS and the dominant (left) PMd for arbitrary visuo-motor mapping when grasping and transporting objects with the dominant (right) hand. Healthy right-handed subjects were asked to reach for and pincer-grasp a neutral coloured cube with the right hand. Visual feedback was controlled with shutter glasses and allowed only at two different time frames at movement onset and 500 ms after movement onset (at the time of peak grasp aperture). Within the course of the reach-to-grasp movement the colour of the cube changed and visual feedback
Human grasping involves a widespread parieto-frontal cortical motor network including the aIPS (putative homologue of the macaque anterior intraparietal area), the PMd, the ventral premotor cortex (PMv) and the primary motor cortex (M1) (Castiello, 2005). There is little data on the anatomical correlates of visuo-motor mapping during reach-to-grasp and transport movements. In healthy humans the cortical correlates of arbitrary visuo-motor mapping during grasping tasks have been probed by creating virtual cortical lesions using TMS (Chouinard et al., 2003, 2005; Davare et al., 2006; Rice et al., 2006; Tunik et al., 2005). Neurons within the aIPS are recruited for somato-sensory object discrimination when grasping objects of different shape (Binkofski et al., 1998; Grefkes and Fink, 2005). A virtual lesion of the left (dominant) aIPS in healthy righthanded subjects, induced by TMS within 65 ms after object perturbation, disrupted the online detection and correction of right hand grasp of an object that unexpectedly changes its size (Rice et al., 2007; Tunik et al., 2005). TMS applied over the left aIPS at the time of peak grasp aperture (150–250 ms prior to object contact), but not over the vertex (control stimulation), left M1 or left PMv, in right-handed subjects grasping and lifting two different masses of identical visual appearance in random order with the right hand disrupted the reactive adjustment of grip force to the novel mass of the object at hand (Dafotakis et al., 2008). A simultaneous bilateral virtual lesion of the human aIPS induced by TMS hampered the exact shaping of the right hand when grasping for an object with constant physical properties in right-handed subjects (Davare et al., 2007; Rice et al., 2007). In contrast, unilateral TMS application over either the left or right aIPS in right-handed subjects grasping for a known object with constant physical properties did not interfere with the exact shaping of the hands (Davare et al., 2007; Tunik et al., 2005). Taken together these data suggest a dual role of aIPS for (i) the integration of intrinsic object properties into motor planning and (ii) rapid online detection of changes of mechanical object features and correction of the ongoing motor action. The critical role of the aIPS for the dynamic online detection and integration of the relevant intrinsic object features into central motor control programs during grasping make an involvement of this area in arbitrary visuo-motor mapping very likely. Within the context of the present data aIPS appears to participate in visuo-motor mapping by rapid integration of the colour of the cube to be matched with the corresponding action based on learned associations as provided by PMd.
M. Taubert et al. / Clinical Neurophysiology 121 (2010) 408–413
4.2. The role of the PMd for arbitrary visuo-motor mapping The PMd is critically involved in the context-specific selection of appropriate motor commands based on arbitrary visuo-motor mappings with external visual cues (Petrides, 1982, 1985). A period of neuro-navigated inhibitory repetitive TMS over the dominant (left) PMd, but not over the left M1, left PMv or the left occipital cortex, in right-handed healthy subjects disturbed the selection of appropriate grip forces to lift a given mass based on learned mappings between arbitrary colours and object masses at both hands (Chouinard et al., 2005; Nowak et al., 2009). This emphasizes the unique role of PMd in the selection of appropriate motor commands based on previously learned visuo-motor mappings with arbitrary visual cues during object manipulation. Our data extend these prior findings by demonstrating that interference with neural processing in PMd during task performance hampers online visuo-motor mapping during grasping. Within this context, aIPS may provide the relevant information related to environmental sensory cues to be matched with the corresponding action as selected by PMd. 4.3. A serial involvement of aIPS and PMd in visuo-motor mapping An interesting observation within the present set of experiments was a time-dependent differential involvement of left aIPS and PMd in the processing of visuo-motor mapping between colour cues and actions during grasping. TMS was delivered at two different time periods of 200 ms duration: TMS 1 during the second shutter glasses opening when the colour cue was presented and TMS 2 after cue presentation but before object contact. TMS over left aIPS at TMS 1, but not at TMS 2, delayed the times to wrist rotation (experiment 1) and hand transport (experiment 2). In contrast, TMS over left PMd at TMS 2, but not at TMS 1, delayed the times to wrist rotation (experiment 1) and hand transport (experiment 2). Note that the time course of cue presentation and visuomotor mapping was different in experiments 1 and 2. In experiment 1, subjects were forced to perform a rapid mapping between colour cue and wrist rotation prior to object contact. In experiment 2, the mapping between colour and hand transport should be performed after object contact. Remarkably, the disturbing effect of TMS was evident in both experiments, indicating that sequential neural processing within aIPS and PMd takes place around the time of cue recognition. The data fit well with the concept that neural processing in aIPS and PMd is serially recruited during visuo-motor mapping when grasping. The fact that TMS of aIPS disturbed visuo-motor mapping when applied during cue presentation points to the essential role of this cortical area for the rapid online detection and integration of sensory information related to mechanical object features (Rice et al., 2006, 2007; Tunik et al., 2005). Interference of TMS with neural processing in PMd after cue presentation highlights the notion of an essential role of this area for the matching between a cue and an intended action and/or action selection. 5. Conclusions We show, for the first time, that the dominant (left) aIPS and PMd are differentially recruited in a serial time locked fashion dur-
413
ing and after cue presentation, respectively, during the mapping of arbitrary visual colour cues with goal directed grasp and transport movements performed with the dominant (right) hand. Acknowledgements The present project was supported by the ‘‘Arbeitskreis Botulinumtoxin” of the Deutsche Gesellschaft für Neurologie (DGN). References Binkofski F, Buccino G, Posse S, Seitz RJ, Rizzolatti G, Freund H. A fronto-parietal circuit for object manipulation in man: evidence from an fMRI-study. Eur J Neurosci 1999;11:3276–86. Binkofski F, Dohle C, Posse S, Stephan KM, Hefter H, Seitz RJ, et al. Human anterior intraparietal area subserves prehension: a combined lesion and functional imaging study. Neurology 1998;50:1253–9. Castiello U. The neuroscience of grasping. Nat Rev Neurosci 2005;6:726–36. Chouinard PA, Leonard G, Paus T. Role of the primary and dorsal premotor cortices in the anticipation of forces during object lifting. J Neurosci 2005;25:2277–84. Chouinard PA, Van Der Werf YD, Leonard G, Paus T. Modulating neural networks with transcranial magnetic stimulation applied over the dorsal premotor and primary motor cortices. J Neurophysiol 2003;90:1071–83. Dafotakis M, Sparing R, Fink GR, Nowak DA. Differential roles of the ventral premotor cortex and anterior intraparietal area for predictive and reactive scaling of grip force. Brain Res 2008;1228:73–80. Davare M, Andres M, Clerget E, Thonnard JL, Olivier E. Temporal dissociation between hand shaping and grip force scaling in the anterior intraparietal area. J Neurosci 2007;27:3974–80. Davare M, Andres M, Cosnard G, Thonnard JL, Olivier E. Dissociating the role of the ventral and dorsal premotor cortex in precision grasping. J Neurosci 2006;26:2260–8. Grefkes C, Fink GR. The functional organization of the intraparietal sulcus in humans and monkeys. J Anat 2005;207:3–17. Murata A, Gallese V, Luppino G, Kaseda M, Sakata H. Selectivity for the shape, size, and orientation of objects for grasping in neurons of macaque parietal area AIP. J Neurophysiol 2000;83:2580–601. Nixon PD, McDonald KR, Gough PM, Alexander IH, Passingham RE. Cortico-basal ganglia pathways are essential for the recall of well-established visuomotor associations. Eur J Neurosci 2004;20:3165–78. Nowak DA, Berner J, Herrnberger B, Kammer T, Groen G, Schönfeldt-Lecuona C. Continuous theta-burst stimulation over the dorsal premotor cortex interferes with associative learning during object lifting. Cortex 2009;45:473–82. Oldfield RC. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 1971;9:97–113. Petrides M. Deficits on conditional associative-learning tasks after frontal and temporal-lobe lesions in man. Neuropsychologia 1985;23:601–14. Petrides M. Motor conditional associative-learning after selective prefrontal lesions in the monkey. Behav Brain Res 1982;5:407–13. Rice NJ, Tunik E, Cross ES, Grafton ST. On-line grasp control is mediated by the contralateral hemisphere. Brain Res 2007;1175:76–84. Rice NJ, Tunik E, Grafton S. The anterior intraparietal sulcus mediates grasp execution, independent of requirement to update: new insights from transcranial magnetic stimulation. J Neurosci 2006;26:8176–82. Stokes MG, Chambers CD, Gould IC, Henderson TR, Janko NE, Allen NB. Simple metric for scaling motor threshold based on scalp-cortex distance. application to studies using transcranial magnetic stimulation. J Neurophysiol 2005;94:4520–7. Tunik E, Frey SH, Grafton S. Virtual lesions of the anterior intraparietal area disrupt goal-dependent on-line adjustments of grasp. Nat Neurosci 2005;8:505–11. Werhahn KJ, Kunesch E, Noachtar S, Benecke R, Classen J. Differential effects on motorcortical inhibition induced by blockade of GABA uptake in humans. J Physiol 1999;517:591–7. Wise SP, Murray EA. Arbitrary associations between antecedents and actions. Trends Neurosci 2000;23:271–6.
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Inhibition of the anterior intraparietal area and the dorsal premotor cortex interfere with arbitrary visuo-motor mapping Marco Taubert a,b,d, Manuel Dafotakis a, Roland Sparing c, Simon Eickhoff a, Siegfried Leuchte d, Gereon R. Fink a,c, Dennis A. Nowak a,c,e,* a
Institute of Neuroscience and Medicine (INB3), Cognitive Neurology Section, Research Centre Jülich, Jülich, Germany Max Planck Institute for Human Cognitive and Brain Sciences, Department of Cognitive Neurology, Leipzig, Germany Department of Neurology, University of Cologne, Cologne, Germany d Department of Sport Science, Martin-Luther-University Halle-Wittenberg, Halle, Germany e Neurologische Fachklinik Kipfenberg, Kipfenberg, Germany b c
a r t i c l e
i n f o
Article history: Accepted 8 November 2009 Available online 9 December 2009 Keywords: Transcranial magnetic stimulation Dorsal premotor cortex Anterior intraparietal sulcus
a b s t r a c t Objective: The contribution of the human anterior intraparietal area and the dorsal premotor cortex to arbitrary visuo-motor mapping during grasping were tested. Methods: Trained right-handed subjects reached for and pincer-grasped a cube with the right hand in the absence of visual feedback after the cube location had been displayed for 200 ms. During the reaching movements, the colour of the cube changed and visual feedback about the change of colour was provided for 100 ms at 500 ms after movement onset (at the time of peak grasp aperture). Depending on colour, subjects were instructed to either pincer-grasp the cube in a horizontal or vertical grasp position with the latter necessitating wrist rotation (experiment 1) or to pincer-grasp and transport the cube to either a left or right target position (experiment 2). Within two consecutive 200 ms time windows (TMS 1 and 2) starting 500 ms and 700 ms after movement onset, respectively, double pulses of supra-threshold transcranial magnetic stimulation (inter-stimulus interval: 100 ms) were delivered over (i) the left primary motor cortex (90° vertically angulated coil position, control stimulation), (ii) the left dorsal premotor cortex or (ii) the left anterior intraparietal area. Results: Compared to control stimulation, stimulation of the anterior intraparietal area, but not of the dorsal premotor cortex, at TMS 1 delayed the times to wrist rotation (experiment 1) and hand transport (experiment 2). Compared to control stimulation, stimulation of the dorsal premotor cortex, but not of the anterior intraparietal area, at TMS 2 delayed both wrist rotation (experiment 1) and hand transport (experiment 2). Conclusions: We contend that the anterior intraparietal area and the dorsal premotor cortex are both involved albeit at different phases during the mapping of arbitrary visual cues with goal directed grasp and transport movements. Significance: These data add to the current understanding of how human cortical areas work in concert during manual activities. Ó 2009 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction Reaching for and grasping an object is part of our daily motor repertoire and is usually generated effortlessly despite a huge variety of potential sources of error. Errors may result from sudden
* Corresponding author. Address: Klinik Kipfenberg, Neurochirurgische und Neurologische Fachklinik, Kindinger Strasse 13, D-85110 Kipfenberg, Germany. Tel.: +49 (0) 8465 175 100; fax: +49 (0) 8465 175 184. E-mail addresses: [email protected], dennis.nowak@ uk-koeln.de (D.A. Nowak).
contextual changes in the environment or from intrinsic noise within the sensory-motor system causing inaccurate movement execution. The accuracy of goal-directed motor performance depends on rapid detection and correction of these errors necessitating online comparison between the issued motor command, current sensory feedback and the specific aim of the intended action. One key issue underlying goal-directed interactions with the environment is the ability to form rapid associations between environmental cues and intended actions. The rapid matching of environmental cues with purposeful actions is referred to as arbitrary visuo-motor mapping (Wise and Murray, 2000).
1388-2457/$36.00 Ó 2009 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.clinph.2009.11.011
M. Taubert et al. / Clinical Neurophysiology 121 (2010) 408–413
Human reaching and grasping seem to involve a widespread parieto-frontal cortical motor network including the anterior intraparietal sulcus (aIPS, putative homologue of macaque anterior intraparietal area), the dorsal premotor cortex (PMd), the ventral premotor cortex and the primary motor cortex (for a recent review see Castiello, 2005). To date, however, less is known about the neural correlates involved in arbitrary visuo-motor mapping. The present study focused on the contributions of the aIPS and PMd to arbitrary visuo-motor mapping during grasping movements. Electrophysiological recordings in the macaque indicate that neurons within the anterior intraparietal area are recruited for visual and/or tactile object discrimination and are preferentially activated for various hand configurations when grasping objects of different shape (Grefkes and Fink, 2005; Murata et al., 2000). Humans with lesions of the aIPS show difficulties to shape their hand according to the intrinsic object characteristics during visually guided reach-to-grasp movements (Binkofski et al., 1998). A virtual lesion of the left (dominant) aIPS in healthy right-handed subjects, induced by transcranial magnetic stimulation (TMS), disrupts the online correction of right hand grasp formation during visually guided reach-to-grasp movements toward an object that unexpectedly changes its size (Tunik et al., 2005; Rice et al., 2006). Thus, aIPS seems to be involved in at least two different processes: (i) the transformation of intrinsic object properties, e.g., size and shape, into a motor plan, and (ii) the online detection of a mismatch between the intended action and environmental context that is necessary to correct grasp formation during visually guided grasping. Lesions of the PMd may result in impairments to reproduce previously acquired visuo-motor mappings or to learn novel mappings, such as linking an arbitrary colour cue to a grasping task, both in macaques and humans (Petrides, 1982, 1985; Nixon et al., 2004). TMS studies have shown that the left (dominant) PMd is essential for the predictive scaling of grasping forces based on learned associations between arbitrary colour cues and the mass of an object to be lifted with either the right or left hand in right-handed subjects (Chouinard et al., 2005; Nowak et al., 2009). Together, these studies indicate that PMd is involved in the context-specific selection of appropriate motor commands based on arbitrary visuo-motor mappings with external cues during grasping. The specific role of the aIPS for (i) the integration of the relevant object features into central motor control programs and (ii) the dynamic online detection of changes in intrinsic object characteristics during grasping (Tunik et al., 2005) make an involvement of aIPS in arbitrary visuo-motor mapping very likely. One putative contribution of aIPS is that it is involved in the online detection and/or integration of arbitrary visual stimuli to be matched with particular actions based on learned associations as provided by PMd. Within this context, aIPS and PMd may be differentially involved at consecutive time intervals in the process of arbitrary visuo-motor mapping during grasping. To test this hypothesis, trained right-handed subjects reached for and pincer-grasped a cube in the absence of visual feedback after the location of the cube had been displayed for 200 ms. During the reach movement the colour of the cube changed and visual feedback about the current cube colour was provided for 100 ms at the time of peak grasp aperture. Depending on the colour, subjects either had to grasp the cube in a horizontal or vertical pincer-grasp position with the latter necessitating hand rotation (experiment 1) or to grasp and transport the cube to either a left or right target position (experiment 2). Within either 0–200 ms (TMS 1) or 200–400 ms (TMS 2) after peak grasp aperture TMS double pulses (inter-stimulus interval: 100 ms) were delivered over (i) the left primary motor cortex (M1, 90° vertically angulated coil position, control stimulation), (ii) the left PMd or (iii) the left aIPS. We hypothesized that, compared to control stimulation over M1, TMS over aIPS and PMd interfered with
409
arbitrary visuo-motor mapping at different time intervals, respectively. 2. Methods 2.1. Participants Ten right-handed (Oldfield, 1971) male subjects (mean age ± SD: 30 ± 7 years) with normal or corrected to normal vision participated. Seven subjects participated in experiment 1; eight subjects participated in experiment 2. Five subjects participated in both experiments. All subjects were naïve to the specific purpose of the study and provided written informed consent. Subjects were screened for neurological disorders and contraindications for TMS. Each subject was able to detect and discriminate the different colours (red, white and black) of the cubes used in experiments 1 and 2. The experimental procedures were approved by the local Ethics committee. 2.2. Neuronavigation procedure A T1-weighted, high-resolution three-dimensional volumetric magnetic resonance imaging (MRI) scan of each subject’s brain was acquired (3 T, Trio, Siemens, Erlangen, Germany; 176 contiguous 1 mm thick sagittal slices). The subject’s head and the individual MRI scan were carefully co-registered in a common reference frame with an infrared optical-tracking neuronavigational system (Polaris System, Northern Digital, Waterloo, Ontario, Canada, and Brainsight software, Rogue Research, Montreal, Canada). The anatomical landmarks used were the tip and the alar wings of the nose, the nibs of the tragus of both ears, and the internal angles of both eyes. The following procedure was used to locate the stimulation site and to monitor the position of the TMS coil over the left M1, left PMd and left aIPS, respectively: First, subject’s structural MRI was transferred into the Montreal Neurological Institute (MNI) coordinate system. Second, probabilistic locations for M1 (x = 31, y = 22, z = 52), PMd (x = 24, y = 3, z = 62) and aIPS (x = 42, y = 39, z = 44) of the left hemisphere were derived based on previous studies (Binkofski et al., 1999; Chouinard et al., 2005; Grefkes and Fink, 2005). Third, the probabilistic locations were transferred to the subject’s brain coordinate space and marked on the three-dimensional MRI scan for stimulation purposes. Control stimulation was performed by positioning the TMS coil in a 90° angulated orientation over the hand area of M1. Single pulse TMS was used to verify correct coil placement with electromyographic activity recorded from the contralateral first dorsal interosseus muscle (see below). The neuronavigational system was then used to navigate the TMS coil over the probabilistic locations marked within the subject’s MRI. The TMS coil was fixed in position over the scalp by a custom made positioning system throughout each session. During the experiments, the centre of the coil was continuously monitored to be over the site of interest and minor deviations were corrected prior to each experimental block. A maximum deviation of 3 mm was tolerated. For stimulation of the PMd and aIPS, the TMS coil was positioned tangential to the surface of the skull with the short axis of the figure-of-eight coil angled 45° relative to the interhemispheric fissure and approximately perpendicular to the central sulcus with the handle pointing lateral and caudal. For control stimulation the TMS coil was placed in 90° vertically angulated coil position at the skull over left M1. That is the magnetic coil was placed vertically on the scalp so that the lateral aspect of the coil touched the scalp and the magnetic field ran tangentially to the scalp. This positioning did not interfere with neural processing within M1. Head movements were restricted by the use of a chin rest and two additional positioning frames.
410
M. Taubert et al. / Clinical Neurophysiology 121 (2010) 408–413
2.3. Transcranial magnetic stimulation (TMS) TMS was performed using a 70-mm figure-of-eight coil and two Magstim 200 connected via a BiStim Module (Magstim Company, Dyfed, UK). The following procedure was performed to obtain the resting motor threshold for each subject. The coil was placed tangentially over the left M1 at the optimal site to elicit a response of the right first dorsal interosseus muscle. The optimal site was defined as the location where stimulation at a slightly supra-threshold intensity elicited the largest motor evoked potential. Electromyographic activity was recorded using silver–silver–chloride electrodes positioned in a belly-tendon technique on the skin overlying the right first dorsal interosseus muscle. The electromyographic signal was amplified, filtered (50–2000 Hz) and digitized at a sampling rate of 5000 Hz. The resting motor threshold was defined for each participant as the lowest stimulator output intensity that elicited motor evoked potentials with peak-to-peak amplitude of at least 50 lV in at least 5 of 10 trials. Two single pulses of TMS (inter-stimulus interval of 100 ms) were applied at an intensity of 100% of the resting motor threshold over M1 and PMd, but 120% of the resting motor threshold over aIPS as the latter is located in the intraparietal sulcus and greater stimulation intensities are necessary to reach the aIPS (Stokes et al., 2005). The application of TMS pulses at intervals of 100 ms induces long interval cortical inhibition (LICI), which is induced by an enhancement of GABA-B receptor activity (Werhahn et al., 1999). 2.4. Kinematic motion analysis Reach-to-grasp movements were recorded using an ultrasonic motion measurement system (CMS 20S, Zebris, Isny, Germany). The system uses three microphones to continuously calculate the three-dimensional spatial positions of small ultrasound emitting markers (diameter: 5 mm) attached via flexible cables to the moving segments of the upper limb. Three position markers were fixed to the styloid process of the radius, the dorsal tip of the index finger and the dorsal tip of the thumb. The position markers placed on the dorsal tips of the index finger and thumb allowed for the registration of the opening and closure of the pincer-grasp. Spatial coordinates of each position marker were sampled at a frequency of 100 Hz and a spatial resolution of 0.1 mm. 2.5. Experimental procedure Subjects were seated in front of a table with their chin positioned in a chin rest to avoid head movements. Subjects wore headphones providing white noise thereby eliminating auditory information during task performance. At the beginning of each trial subjects placed the wrist of their right hand on a start button and pincer-grasped a piece of tape between thumb and index finger to standardize initial hand position. Subjects were instructed to reach for and pincer-grasp a cube positioned on a turning table 42 cm away from the start button and 20 cm below shoulder level between the index finger and thumb. Two cubes of equal dimensions (2 2 2 cm) were fixed on a turning table mounted on the shaft of a motor (model TMCM; Trinamic, Hamburg, Germany) separated by a vertical wall in the middle of the table. The cubes had different shapes in experiments 1 and 2. In experiment 1, subjects had to grasp the cube between the index finger and thumb in a horizontal or vertical grasp (Fig. 1). During a vertical grasp the index finger should be placed on the top plane and the thumb on the bottom plane of the cube. Thus, the cube in experiment 1 was positioned in 30° angulation on the turning table (see Fig. 1). In experiment 2, subjects always grasped the cube in a horizontal grasp. Thus, the cube was positioned on its bottom plane on the turning table in experiment 2. In both experiments, two similar shaped,
but different-coloured, cubes were placed on each side of the wall so that subjects were able to see the cube on the front side of the turning table. The motor of the turning table was programmed to turn the table for 180° after movement initiation (release of the start button) to present the cube on the other side of the wall (time for motor to complete rotation: 100 ms). Visual feedback was controlled by liquid crystal shutter glasses (Plato System; Translucent Technologies, Toronto, Ontario, Canada). The shutter glasses (time to switch between light scattering opaque and transparent state: 4 ms) remained opaque during each trial except for a 200 ms window prior to movement onset, which served as a go-signal to initiate the movement and a 100 ms window initiated 500 ms after movement onset (corresponding to the time of peak grasp aperture) (see Fig. 1). Rotation of the turning table presented the cube on the back side of the wall, which was visible within the second 100 ms window the shutter glasses were transparent. The shutter glasses then turned opaque again for the remainder of the trial. In experiment 1 (n = 7), subjects saw a red cube (on the front side of the turning table) during the first time window and a different-coloured, white or black, cube (on the back side of the turning table) during the second time window the shutter glasses turned transparent 500 ms after movement onset. The colour of the cube indicated that subjects had to pincer-grasp the cube in a horizontal grasp along its lateral surfaces (black cube) or vertical grasp along its upper and lower surfaces (white cube). Grasping the cube in a vertical pincer-grasp required a rapid wrist rotation before object contact. In experiment 2 (n = 8), similar to the first experiment, subjects saw a red cube (on the front side of the turning table) in the first time window and different-coloured, white or black, cube (on the back side of the turning table) in the second time window. Here, the colour of the cube indicated that the cube had to be grasped in a horizontal pincer-grasp at its lateral surfaces and rapidly transported and placed into one of two boxes on the left (white cube) or right side (black cube) of the turning desk (distance 15 cm). Note that subjects had to finish the reach-to-grasp movement before transporting the cube to either of the two boxes. During the reach-to-grasp movement in both experiments, double pulse TMS (inter-stimulus interval of 100 ms; Rice et al., 2006) were applied (i) 500–700 ms (TMS 1) and (ii) 700–900 ms (TMS 2) after movement onset. Note that TMS 1 included the period the shutter glasses turned transparent for a second time (see Fig. 1). TMS 1 during the transparent phase of the shutter glasses was thought to interfere with the detection and integration of the arbitrary colour cue. TMS 2 was thought to interfere with the selection of motor commands for wrist rotation (experiment 1) or transport movement (experiment 2) based on arbitrary colour cues. TMS was applied over left M1, PMd and aIPS on separate days. The sequence of TMS application was randomly assigned to each subject. On each day subjects performed 5 blocks of 20 trials. Within each block an equal number of black and white cubes were presented for TMS 1 and TMS 2 in random order. This procedure ensured that subjects were unable to predict cube colour of the subsequent trial. Prior to each experimental session, subjects were trained by performing practice session (4 blocks of 20 trials each) to learn the task requirements. 2.6. Data analysis The X-, Y- and Z-directions of the position marker coordinates refer to the medio-lateral, antero-posterior and vertical directions with regard to the subject performing the task. The software (3DA, MedCom, Munich, Germany) also allowed for the calculation of distances between single position markers. For each reach-tograsp movement, we determined movement onset as the first detectable 5% change of wrist velocity in the Z-direction from base-
M. Taubert et al. / Clinical Neurophysiology 121 (2010) 408–413
411
Fig. 1. Illustration of the experimental paradigms for experiments 1 and 2. Subjects were seated in front of a table. At the beginning of each trial subjects placed the wrist of their right hand on a start button and pincer-grasped a piece of tape between thumb and index finger to standardize initial hand position. Subjects wore headphones which provided white noise during task performance. Subjects were instructed to reach for and pincer-grasp a cube positioned on a turning table between the index finger and thumb. Two cubes of equal dimensions were fixed on a turning table separated by a vertical wall in the middle of the table. The cubes were placed on each side of the wall so that subjects were able to see the cube on the front side of the turning table. The table turned 180° after movement initiation (release of the start button) to present the cube on the other side of the wall. Visual feedback was controlled by liquid crystal shutter glasses. The shutter glasses remained opaque during the trial except for a 200 ms window prior to movement onset, which served as a go-signal to initiate the movement and a 100 ms window initiated 500 ms after movement onset. The colour of the cube displayed at the time of peak grasp aperture indicated to either grasp the cube in a horizontal or vertical position (experiment 1) or grasp and transport the cube to a left or right target position (experiment 2).
line obtained from the marker fixed to the styloid process of the radius. Performance in experiment 1 was quantified by assessing the time (in ms) from the onset of the second shutter glasses opening (500 ms after movement onset) to the onset of wrist rotation. Only the trials in which the wrist was required to rotate from a horizontal to a vertical grasp position were included in the data analysis. The onset of wrist rotation was determined as the first detectable 5% change in the angular velocity of the index finger marker within the X-plane. Performance in experiment 2 was quantified by assessing (i) the time needed (in ms) from the onset of the second shutter glasses opening to object contact (minimal distance between thumb and index finger markers and velocity offset) and (ii) the time (in ms) from object contact to the onset of the transport movement to the left or right box (first detectable 5% change of index finger velocity in the X-direction from baseline). All parameters were averaged across all trials performed by each participant. Mean values were calculated for each subject within both experiments. Repeated measures analyses of variance (ANOVA) with a level of significance of P < 0.05 were calculated to assess the effect of ‘‘stimulation site” (TMS left M1, left PMd and left aIPS) and ‘‘time” (TMS1 and TMS2) on reaction times. Post-hoc t-tests were corrected for multiple testing according to Bonferroni. 3. Results 3.1. Experiment 1 Fig. 2a illustrates average times (+1 SEM) from the onset of visual feedback about cube colour (onset of the second shutter glasses opening, see Fig. 1) to the initiation of wrist rotation. It is evident that when applied over the left aIPS, but not over the left PMd, TMS 1 delayed the times needed to perform the wrist rotation compared to control stimulation over the left M1 (with 90°
angulated coil orientation). In contrast, TMS 2 prolonged the times needed to rotate the hand when applied over the left PMd, but not over the left aIPS, compared to control stimulation. These effects were significant: repeated measures ANOVA revealed no significant effects of the single factors ‘‘stimulation site” or ‘‘time”. However, there was a significant effect of the interaction ‘‘stimulation site” ‘‘time” on times to hand rotation (F1,6 = 14.6; P < 0.001). Post-hoc t-tests showed that for TMS 1 the times needed to rotate the wrist were longer after TMS over the left aIPS compared to stimulation over the left PMd or control stimulation over left M1 (P < 0.01 for each comparison). In addition, for TMS 2 the times needed to rotate the wrist were longer after TMS over the left PMd compared to TMS over the left aIPS or the left M1 (P < 0.01 for each comparison). 3.2. Experiment 2 In experiment 2, the times from the onset of the second shutter glasses opening to object contact were not significantly prolonged by TMS, regardless of the site or time of stimulation. Repeated measures ANOVA detected neither a significant effect of the single factors ‘‘stimulation site” or ‘‘time” nor their interactions on times to object contact. Fig. 2b illustrates average times (+1 SEM) from object contact to the onset of the transport movement of the hand to the left or right box. Fig. 2b shows average times (+1 SEM) from the onset of the second shutter glasses opening to the time of hand transport initiation. Repeated measures ANOVA detected no significant effects of the single factors ‘‘stimulation site” or ‘‘time” on times to hand transport. A significant interaction ‘‘stimulation site” ‘‘time” on times to hand transport (F1,7 = 14.6; P < 0.001) suggested differential effects of stimulation site on times to hand transport depending on the time of stimulation. Pot hoc tests revealed that for TMS 1
412
M. Taubert et al. / Clinical Neurophysiology 121 (2010) 408–413
about the actual colour of the object was displayed for 100 ms at the time of peak grasp aperture. The colour indicated that subjects either had to pincer-grasp the cube in a horizontal or vertical grasp position with the latter necessitating a wrist rotation (experiment 1) or to pincer-grasp and transport the cube to either a left or right target position (experiment 2). TMS was used to modulate neural processing of (i) the left primary motor cortex (90° vertically angulated coil position, control stimulation), (ii) the left PMd or (iii) the left aIPS during (TMS 1) or after (TMS 2) the presentation of the visual colour cue. These differential time points for TMS were introduced to detect potential sequential processing in aIPS and PMd. Stimulation of left aIPS, but not of left PMd, disturbed the times to the onset of wrist rotation (experiment 1) and onset of hand transport (experiment 2) when given at TMS 1. In contrast, stimulation of left PMd, but not left aIPS, at TMS 2 delayed the times to the initiation of hand rotation (experiment 1) and hand transport (experiment 2). These data add new evidence to the functional temporal coupling between aIPS and PMd during visuo-motor mapping.
4.1. The role of the aIPS for arbitrary visuo-motor mapping
Fig. 2. (a) Average times (+1 SEM) from the onset of the second time window of shutter glasses opening to the initiation of wrist rotation (experiment 1, N = 7) and (b) average times from object contact to the initiation of the wrist transport movement (experiment 2, N = 8). The times to the initiation of wrist rotation (experiment 1) and the times to the initiation of wrist transport (experiment 2) were significantly longer following early TMS 1 over left aIPS compared to TMS over left PMd or control stimulation over left M1. In contrast, the times to the initiation of wrist rotation (experiment 1) and the times to the initiation of wrist transport (experiment 2) were significantly longer following late TMS 2 over left PMd compared to TMS over left aIPS or control stimulation over left M1. For the definition of the time points of TMS application within the time course of the experiment refer to Fig. 1. Significant differences are indicated (**P < 0.01).
times from the second shutter glasses opening to initiation of hand transport were longer when TMS was applied over the left aIPS, compared to stimulation over the left PMd or control stimulation over left M1 (P < 0.01 for each comparison). In contrast, for TMS 2 times to hand transport onset were longer when TMS was applied over the left PMd, compared to stimulation over the left aIPS or control stimulation over the left M1 (P < 0.01 for each comparison). 4. Discussion The present experiments were designed to investigate the contribution of the (dominant) left aIPS and the dominant (left) PMd for arbitrary visuo-motor mapping when grasping and transporting objects with the dominant (right) hand. Healthy right-handed subjects were asked to reach for and pincer-grasp a neutral coloured cube with the right hand. Visual feedback was controlled with shutter glasses and allowed only at two different time frames at movement onset and 500 ms after movement onset (at the time of peak grasp aperture). Within the course of the reach-to-grasp movement the colour of the cube changed and visual feedback
Human grasping involves a widespread parieto-frontal cortical motor network including the aIPS (putative homologue of the macaque anterior intraparietal area), the PMd, the ventral premotor cortex (PMv) and the primary motor cortex (M1) (Castiello, 2005). There is little data on the anatomical correlates of visuo-motor mapping during reach-to-grasp and transport movements. In healthy humans the cortical correlates of arbitrary visuo-motor mapping during grasping tasks have been probed by creating virtual cortical lesions using TMS (Chouinard et al., 2003, 2005; Davare et al., 2006; Rice et al., 2006; Tunik et al., 2005). Neurons within the aIPS are recruited for somato-sensory object discrimination when grasping objects of different shape (Binkofski et al., 1998; Grefkes and Fink, 2005). A virtual lesion of the left (dominant) aIPS in healthy righthanded subjects, induced by TMS within 65 ms after object perturbation, disrupted the online detection and correction of right hand grasp of an object that unexpectedly changes its size (Rice et al., 2007; Tunik et al., 2005). TMS applied over the left aIPS at the time of peak grasp aperture (150–250 ms prior to object contact), but not over the vertex (control stimulation), left M1 or left PMv, in right-handed subjects grasping and lifting two different masses of identical visual appearance in random order with the right hand disrupted the reactive adjustment of grip force to the novel mass of the object at hand (Dafotakis et al., 2008). A simultaneous bilateral virtual lesion of the human aIPS induced by TMS hampered the exact shaping of the right hand when grasping for an object with constant physical properties in right-handed subjects (Davare et al., 2007; Rice et al., 2007). In contrast, unilateral TMS application over either the left or right aIPS in right-handed subjects grasping for a known object with constant physical properties did not interfere with the exact shaping of the hands (Davare et al., 2007; Tunik et al., 2005). Taken together these data suggest a dual role of aIPS for (i) the integration of intrinsic object properties into motor planning and (ii) rapid online detection of changes of mechanical object features and correction of the ongoing motor action. The critical role of the aIPS for the dynamic online detection and integration of the relevant intrinsic object features into central motor control programs during grasping make an involvement of this area in arbitrary visuo-motor mapping very likely. Within the context of the present data aIPS appears to participate in visuo-motor mapping by rapid integration of the colour of the cube to be matched with the corresponding action based on learned associations as provided by PMd.
M. Taubert et al. / Clinical Neurophysiology 121 (2010) 408–413
4.2. The role of the PMd for arbitrary visuo-motor mapping The PMd is critically involved in the context-specific selection of appropriate motor commands based on arbitrary visuo-motor mappings with external visual cues (Petrides, 1982, 1985). A period of neuro-navigated inhibitory repetitive TMS over the dominant (left) PMd, but not over the left M1, left PMv or the left occipital cortex, in right-handed healthy subjects disturbed the selection of appropriate grip forces to lift a given mass based on learned mappings between arbitrary colours and object masses at both hands (Chouinard et al., 2005; Nowak et al., 2009). This emphasizes the unique role of PMd in the selection of appropriate motor commands based on previously learned visuo-motor mappings with arbitrary visual cues during object manipulation. Our data extend these prior findings by demonstrating that interference with neural processing in PMd during task performance hampers online visuo-motor mapping during grasping. Within this context, aIPS may provide the relevant information related to environmental sensory cues to be matched with the corresponding action as selected by PMd. 4.3. A serial involvement of aIPS and PMd in visuo-motor mapping An interesting observation within the present set of experiments was a time-dependent differential involvement of left aIPS and PMd in the processing of visuo-motor mapping between colour cues and actions during grasping. TMS was delivered at two different time periods of 200 ms duration: TMS 1 during the second shutter glasses opening when the colour cue was presented and TMS 2 after cue presentation but before object contact. TMS over left aIPS at TMS 1, but not at TMS 2, delayed the times to wrist rotation (experiment 1) and hand transport (experiment 2). In contrast, TMS over left PMd at TMS 2, but not at TMS 1, delayed the times to wrist rotation (experiment 1) and hand transport (experiment 2). Note that the time course of cue presentation and visuomotor mapping was different in experiments 1 and 2. In experiment 1, subjects were forced to perform a rapid mapping between colour cue and wrist rotation prior to object contact. In experiment 2, the mapping between colour and hand transport should be performed after object contact. Remarkably, the disturbing effect of TMS was evident in both experiments, indicating that sequential neural processing within aIPS and PMd takes place around the time of cue recognition. The data fit well with the concept that neural processing in aIPS and PMd is serially recruited during visuo-motor mapping when grasping. The fact that TMS of aIPS disturbed visuo-motor mapping when applied during cue presentation points to the essential role of this cortical area for the rapid online detection and integration of sensory information related to mechanical object features (Rice et al., 2006, 2007; Tunik et al., 2005). Interference of TMS with neural processing in PMd after cue presentation highlights the notion of an essential role of this area for the matching between a cue and an intended action and/or action selection. 5. Conclusions We show, for the first time, that the dominant (left) aIPS and PMd are differentially recruited in a serial time locked fashion dur-
413
ing and after cue presentation, respectively, during the mapping of arbitrary visual colour cues with goal directed grasp and transport movements performed with the dominant (right) hand. Acknowledgements The present project was supported by the ‘‘Arbeitskreis Botulinumtoxin” of the Deutsche Gesellschaft für Neurologie (DGN). References Binkofski F, Buccino G, Posse S, Seitz RJ, Rizzolatti G, Freund H. A fronto-parietal circuit for object manipulation in man: evidence from an fMRI-study. Eur J Neurosci 1999;11:3276–86. Binkofski F, Dohle C, Posse S, Stephan KM, Hefter H, Seitz RJ, et al. Human anterior intraparietal area subserves prehension: a combined lesion and functional imaging study. Neurology 1998;50:1253–9. Castiello U. The neuroscience of grasping. Nat Rev Neurosci 2005;6:726–36. Chouinard PA, Leonard G, Paus T. Role of the primary and dorsal premotor cortices in the anticipation of forces during object lifting. J Neurosci 2005;25:2277–84. Chouinard PA, Van Der Werf YD, Leonard G, Paus T. Modulating neural networks with transcranial magnetic stimulation applied over the dorsal premotor and primary motor cortices. J Neurophysiol 2003;90:1071–83. Dafotakis M, Sparing R, Fink GR, Nowak DA. Differential roles of the ventral premotor cortex and anterior intraparietal area for predictive and reactive scaling of grip force. Brain Res 2008;1228:73–80. Davare M, Andres M, Clerget E, Thonnard JL, Olivier E. Temporal dissociation between hand shaping and grip force scaling in the anterior intraparietal area. J Neurosci 2007;27:3974–80. Davare M, Andres M, Cosnard G, Thonnard JL, Olivier E. Dissociating the role of the ventral and dorsal premotor cortex in precision grasping. J Neurosci 2006;26:2260–8. Grefkes C, Fink GR. The functional organization of the intraparietal sulcus in humans and monkeys. J Anat 2005;207:3–17. Murata A, Gallese V, Luppino G, Kaseda M, Sakata H. Selectivity for the shape, size, and orientation of objects for grasping in neurons of macaque parietal area AIP. J Neurophysiol 2000;83:2580–601. Nixon PD, McDonald KR, Gough PM, Alexander IH, Passingham RE. Cortico-basal ganglia pathways are essential for the recall of well-established visuomotor associations. Eur J Neurosci 2004;20:3165–78. Nowak DA, Berner J, Herrnberger B, Kammer T, Groen G, Schönfeldt-Lecuona C. Continuous theta-burst stimulation over the dorsal premotor cortex interferes with associative learning during object lifting. Cortex 2009;45:473–82. Oldfield RC. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 1971;9:97–113. Petrides M. Deficits on conditional associative-learning tasks after frontal and temporal-lobe lesions in man. Neuropsychologia 1985;23:601–14. Petrides M. Motor conditional associative-learning after selective prefrontal lesions in the monkey. Behav Brain Res 1982;5:407–13. Rice NJ, Tunik E, Cross ES, Grafton ST. On-line grasp control is mediated by the contralateral hemisphere. Brain Res 2007;1175:76–84. Rice NJ, Tunik E, Grafton S. The anterior intraparietal sulcus mediates grasp execution, independent of requirement to update: new insights from transcranial magnetic stimulation. J Neurosci 2006;26:8176–82. Stokes MG, Chambers CD, Gould IC, Henderson TR, Janko NE, Allen NB. Simple metric for scaling motor threshold based on scalp-cortex distance. application to studies using transcranial magnetic stimulation. J Neurophysiol 2005;94:4520–7. Tunik E, Frey SH, Grafton S. Virtual lesions of the anterior intraparietal area disrupt goal-dependent on-line adjustments of grasp. Nat Neurosci 2005;8:505–11. Werhahn KJ, Kunesch E, Noachtar S, Benecke R, Classen J. Differential effects on motorcortical inhibition induced by blockade of GABA uptake in humans. J Physiol 1999;517:591–7. Wise SP, Murray EA. Arbitrary associations between antecedents and actions. Trends Neurosci 2000;23:271–6.