Supplementary Figure 1

Transcription

1 Supplementary Figure 1 Left aspl Right aspl Detailed description of the fmri activation during allocentric action observation in the aspl. Averaged activation (N=13) during observation of the allocentric clips in the aspl ROIs. The hand preference (for the ipsilateral hand) can be seen in both visual fields. Thus, right hand actions generate greater activation in the right aspl than left hand actions, in both visual fields (and vice versa in the left aspl). Indeed, an analysis of variance of the beta weights assigned to each condition during observation of the four allocentric conditions in the ROIs revealed that there was no significant interaction between the visual field and the hand identity (F 1,12 =0.4 in the left aspl and F 1,12 =0.18 in the right aspl). A significant hand identity effect was found in both hemispheres (F 1,12 = 27.43, P<0.001 in the left aspl, and F 1,12 =11.88, P<0.005 in the right aspl). Additionally, significant visual field effect was found in the right aspl (F 1,12 =7.89, P<0.02).

2 Supplementary Figure 2 Hand identity effect in the aspl ROIs can be seen in the two visual fields Comparison between the activation during observation of the allocentric conditions in the ipsilateral and contralateral visual fields, pooled across hemispheres. Each condition was named according to the seen hand (ipsilateral/contralateral) and the visual field (ipsilateral/contralateral). In each subject, activation was normalized by subtracting the mean activation level during all 4 allocentric conditions. This was done separately for the right aspl and the left aspl. The results show the averaged activation (N=13) across the two hemispheres. Although on average there is greater activation in the contralateral visual field (blue bars) than in the ipsilateral one (orange bars) (i.e. a visual field effect), the differences between the activation elicited by viewing the ipsilateral hand and the contralateral one are clear and statistically significant in both visual fields (paired t-test, P<0.04 for the ipsilateral field, P<0.001 in the contralateral field).

3 Supplementary Figure 3 Action observation effects Direct contrast between the object manipulation clips presented in the right and left visual field revealed a strong visual field effect in the occipito-temporal cortex. Thus, clips that were presented in the contralateral visual field elicited significantly higher activation than clips that were presented in the ipsilateral visual field. Contralateral preference to the location of the clip in the visual field is shown in green. The viewpoint effect was calculated by comparing the activation during observation of the hand actions shown from an egocentric and allocentric points of view. Preference for the allocentric clips can be seen in the lateral occipital complex of the left hemisphere (orange clusters). No significant preference for the egocentric clips was found. Hand identity effect was calculated by comparing the activation during observation of left hand action and right hand action. This contrast revealed a small cluster in the left frontal cortex. This cluster demonstrated a preference for right hand action over left hand actions. No significant opposite preference was found. Finally, mirror like representation, i.e, differential activation during right egocentric and left allocentric clips (compared to left egocentric and right allocentric clips), can be seen in the superior parietal lobule (purple clusters). Consistent with figure 1 in the main text, the purple clusters in the left hemisphere represent preference for the right egocentric and left allocentric clips (compared with the left egocentric and right allocentric clips) and the cluster in the right SPL represents the opposite preference.

4 Supplementary Figure 4 Representation of observed hand actions in the primary motor cortex. One concern is that the differential activation in the aspl stems from unconscious execution of movements during observation of the clips. To address this issue, we measured the activation in the primary motor cortex during observation of hand actions. First, we identified the primary motor cortex of the 9 subjects that participated in the somatomotor mapping experiment by looking for voxels that show significantly higher activation during execution of actions with the contralateral hand than the ipsilateral hand in the central sulcus (P<0.001). Then, we measured the activation in this ROI during the second experiment, which was consisted of imitation and observation conditions. The only significant activation was found in the contralateral (left) M1 during imitation of action (one sample t-test, P<0.001). During the observation of action, the activation in M1 did not differ significantly from zero. This clear distinction makes it extremely unlikely that the observation epochs were unconscious imitation epochs.

5 Methods Subjects Thirteen right handed volunteers without neurological, psychiatric or visual deficits history (seven women and seven men, ages 25 49) gave their written consent and participated in the present fmri experiments. The Hadassah Ein Karem Medical Center Ethics Committee approved the experimental procedure. MRI acquisition The BOLD fmri measurements were performed in a whole-body 1.5-T Magnetom Avanto SIMENS scanner. The functional MRI protocols were based on a multi-slice gradient echo-planar imaging and a standard head coil. The functional data were obtained under the optimal timing parameters: TR = 3sec, TE = 50ms, flip angle = 90, imaging matrix = 64*64, FOV = 20cm. The 32 slices with slice thickness 3mm (with 0.3mm gap) were oriented in the axial position. The scan covered the whole brain. Experimental setup Footages were taken using a digital camera (Sony trv60e), edited (using the program: "windows movie maker"), and projected via LCD projector (Epson MP 7200, Japan) onto a tangent screen located inside the scanner in front of the subject. Subjects viewed the screen through a tilted mirror. Visual stimuli A set of 32 object manipulation clips, 12 sec long, and 2 visual control clips were shown in this experiment. Right hand clips were composed of twelve ms, black and white footages, of a right hand approaching grasping and releasing an object (total time 12 sec). The footages were taken in two different settings. In the first, the camera was behind the acting hand, and in the second, it was in front of the acting hand. The footages were taken using a set of 20 man-made small graspable objects, such as different cubes, cylinders, pens, scissors, etc. Six to ten footages were taken with each object, showing different grasping movements. Left hand clips were generated by mirror-image transformation of the footages showing right hand actions.

6 In all clips the objects remained stationary, and did not move throughout the grasping movements. In the control "scramble" conditions, a 12-sec object-manipulation clip was decomposed to frames (30 frames a second), every frame was spatially scrambled (36*24 fragments in each frame, ensuring that the hand and the objects could no longer be identified) and then the frames were recomposed to a 12-sec clip. A fixation point appeared in the middle of the screen throughout the experiment. Experimental paradigm: Action observation experiment: The experiment was carried out using a block design format. Ten blocks (consisting of 8 different object grasping conditions, and 2 "scramble" conditions) were interleaved, and were repeated 4 times, in two different runs, using different stimuli, in a counter-balanced manner. Each block lasted 12 s followed by a blank period of 9 s. The first and last blank periods were longer (21 and 15 s, respectively). The clips were centered 16.5 degrees to the right or the left of the fixation point, and subtended 20*15 degrees. Subjects were trained before the experiment to make sure they maintain fixation and understand the instructions. Action observation and imitation experiment: Nine out of the 13 subjects (that participated in the original action observation experiment) did an additional action observation and imitation experiment, as well as a somatomotor mapping run. The experiment was carried out using a block design format. Ten blocks were interleaved, and were repeated 6 times, in two different runs. Each block lasted 12 s followed by a blank period of 9 s. The first and last blank periods were longer (21 and 15 s, respectively). The clips were presented in the middle of the screen and subtended 20*15 degrees. The color of the fixation point indicated the task: red fixation point indicated a perceptual task, in which the subjects had to covertly assess if the current hand action is the same as the previous one. Green fixation point served as a cue for an imitation task the subjects had to imitate the observed grasping movement with their right hand. Due to their supine posture, the subjects could not see their hands. Two control conditions were introduced. In the visual control red fixation the subjects observed a scrambled version of the clips. In the motor execution control, the subjects observed the scrambled version of the clips, and had to open and close their palm according to the dimming of the green

7 fixation point. The fixation point dimmed 12 times during the block, as the required frequency of imitation in the action observation blocks. Somatomotor mapping During this session, subjects had to manipulate a cube with their right or left hand for 12 seconds, according to aural instructions. The subjects focused on a fixation point at the center of the screen throughout the experiment. Due to the supine posture of the subjects, they could not see their actions. Data analysis Data analysis was performed using the BrainVoyager Qx software package (Brain Innovation, Maastricht, The Netherlands, 2000). For each scan, the 2D functional data were examined for motion and signal artifacts. Head motion correction, high-pass temporal smoothing in the frequency domain (cutoff frequency: 3 cycles/ total scan time), and slice scan time correction were applied in order to remove drifts and to improve the signal-to-noise ratio. The complete data set was transformed into Talairach space, Z normalized and concatenated. A general linear model (GLM) approach was used to generate statistical parametric maps; the hemodynamic response function was modeled using a two-gamma hemodynamic response function. Significance levels were calculated, taking into account the probability of a false detection for any given cluster, by requiring that statistically significant activation is seen in contiguous groups of voxels. The implemented method is based on the approach of Forman and colleagues 1 and accomplished by Monte Carlo simulations (using a BV QX plug-in). Across-subject statistical parametric maps (Fig. 1 and 3, Sup. Fig. 1) were calculated using a hierarchical random-effects model analysis 2. The cortical surface was reconstructed from the T1-weighted isovoxel scan. The procedure included segmentation of the white matter using a grow-region function, the smooth covering of a sphere around the segmented region, and the expansion of the reconstructed white matter into the gray matter. Region of interest analysis

8 The multi-subject ROIs were identified in the parietal lobe based on their significant preference for observation of the contralateral hand, when shown from an egocentric point of view (compared to observation of the ipsilateral hand from the same viewpoint, P<0.05, cluster-size correction). The activation time course of individual subjects was obtained from the ROIs. The time course of each run was Z-normalized. For each subject, beta weights assigned to each condition were calculated based on a GLM analysis. These were then averaged across subjects, to yield the shown bar histograms (Fig. 2 and 3, Sup. Fig. 1,2,4). Paired two-tailed t-test was applied to test for significant differences between conditions. ANOVA for repeated measures was applied to calculate the significance of the different effects (hand identity and viewpoint) and the interaction between them. Reference 1. Forman, S. D., Cohen, J. D., Fitzgerald, M., Eddy, W. F., Mintun, M. A., Noll, D. C. Magn. Reson. Med (1995). 2. Friston, K. J., Holmes, A. P., Worsley, K. J. Neuroimage (1999).