Spatial interference during bimanual coordination: differential brain networks associated with control of movement amplitude and direction

Abstract

Bimanual interference emerges when spatial features, such as movement direction or amplitude, differ between limbs, as indicated by a mutual bias of limb trajectories. Although first insights into the neural basis of directional interference have been revealed recently, little is known about the neural network associated with amplitude interference. We investigated whether amplitude versus directional interference activates differential networks. Functional magnetic resonance imaging (fMRI) was applied while subjects performed cyclical, bimanual joystick movements with either the same vs. different amplitudes, directions, or both. The kinematic analysis confirmed that subjects experienced amplitude interference when they moved with different as compared to the same amplitude, and directional interference when they moved along different as compared to the same direction. On the brain level, amplitude and directional interference both resulted in activation of a bilateral superior parietal-premotor network, which is known to contribute to sensorimotor transformations during goal-directed movements. Interestingly, amplitude but not directional interference exclusively activated a bilateral network containing the dorsolateral prefrontal cortex, anterior cingulate, and supramarginal gyrus, which was shown previously to contribute to executive functions. Even though the encoding of amplitude and directional information converged and activated the same neural substrate, our data thus show that additional and partly independent mechanisms are involved in bimanual amplitude as compared to that in directional control.

Figure 1

A: Experimental setup showing exemplary movements along orthogonal directions. B: Amplitude and direction requirements during the four bimanual conditions as symbolized by pictograms. The three lines indicate the amplitude/direction requirements for the first, second, and third interval for the left (left part) and right wrist (right part). The length of the lines reflects whether 30, 60, or 90% of the maximal amplitude were required. C: Exemplary time course of the produced amplitudes and produced directions (D) of a typical subject for the right (gray) and left hand (black) when movements with different amplitudes and different directions were required (Amp&DirInterf condition). The left‐hand amplitude (C, black trace) exhibits clear deviations from the required constant medium amplitude.

Figure 2

Mean amplitude (meanAmp) data and standard errors on group level for conditions requiring the same (A) or different amplitudes (B). Data for movements along the same (black squares) and different directions (gray triangles) are shown for the left and the right wrist as well as the small, medium, and large amplitude requirements (corresponding to 30, 60, and 90% of the individual maximal amplitude) separately. Note that when subjects were to move simultaneously along different amplitudes (B), the mean amplitude of the left wrist exhibited clear modulations, even though subjects were required to move with a constant medium amplitude.

Figure 3

Mean group results and standard errors of the amplitude standard deviation (sdAmp; A), directional error (errorDir; B), and directional standard deviation (sdDir; C). Data are shown for the same vs. different amplitudes (abscissa) and for movements along the same (black squares) vs. different directions (gray triangles). Each data point was yielded by averaging across the three amplitude requirements (i.e., small, medium, and large) and both wrists.

Figure 4

A: Top view of areas exhibiting an amplitude interference main effect (red), a directional interference main effect (blue), or both (purple). B: More detailed view of the interference main effects as in A within the dorsal (upper slice) and ventral premotor cortex (lower slice). C: Network exhibiting a significant amplitude interference × directional interference interaction. Activation is superimposed on top of a rendered brain shown in neurological convention (right is right). [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

Figure 5

Brain slices showing areas that exhibited a significant amplitude interference main effect (A–E), a significant directional interference main effect (F–H), and a significant conjunction effect (I, J). For selected regions line plots are shown displaying the blood oxygenation level‐dependent (BOLD) response in arbitrary units for movements with the same (sAmp) vs. different amplitudes (dAmp) and along the same (black squares) and different (gray triangles) directions. All ordinates of the line plots are scaled to the same range of 1,000 units. Slices are shown in neurological convention (right is right) and coordinates are reported with respect to the Montreal Neurological Institute (MNI) reference brain. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

Figure 6

A–C: Brain slices showing areas that exhibit a significant Amplitude interference × Directional interference interaction effect. The same conventions are used as in Figure 5. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]