Losing dexterity: patterns of impaired coordination of finger movements in musician's dystonia

Abstract

Extensive training can bring about highly-skilled action, but may also impair motor dexterity by producing involuntary movements and muscular cramping, as seen in focal dystonia (FD) and tremor. To elucidate the underlying neuroplastic mechanisms of FD, the present study addressed the organization of finger movements during piano performance in pianists suffering from the condition. Principal component (PC) analysis identified three patterns of fundamental joint coordination constituting finger movements in both patients and controls. The first two coordination patterns described less individuated movements between the "dystonic" finger and key-striking fingers for patients compared to controls. The third coordination pattern, representing the individuation of movements between the middle and ring fingers, was evident during a sequence of strikes with these fingers in controls, which was absent in the patients. Consequently, rhythmic variability of keystrokes was more pronounced during this sequence of strikes for the patients. A stepwise multiple-regression analysis further identified greater variability of keystrokes for individuals displaying less individuated movements between the affected and striking fingers. The findings suggest that FD alters dexterous joint coordination so as to lower independent control of finger movements, and thereby degrades fine motor control.

Figure 1. Time-varying finger joint angles during piano playing.

Time-course profiles of the angular position at the MCP (left), PIP (middle), and DIP (ring) joints of the index (red), middle (green), ring (blue) and little (black) fingers and the corresponding MIDI information (black horizontal bars) in one representative healthy pianist (“Control”) and a pianist with FD at the index finger (“Patient”). x-axis: normalized time (1000 time-points from the key-press of the initial note to the key-release of the final note). The hand in the figure was drawn by the corresponding author.

Figure 2. Maximum joint extension and flexion angles.

Group means of the maximum (top panel) and minimum (bottom panel) angles at the MCP (left), PIP (middle), and DIP (right) joints of the index (I), middle (M), ring (R), and little (L) fingers while the healthy control pianists (white bar) and FD pianists (black bar) were performing a sequence of keystrokes. Negative and positive values indicate flexion and extension, respectively. Error bars indicate one standard deviation within a group. *p < 0.05, **p < 0.01.

Figure 3. Time-varying PC waveforms.

Time-varying waveforms of the first 3 PCs averaged within the controls (left) and within the patients with FD (right). The thick lines and shaded bands indicate the mean and one standard deviation, respectively, within each of the groups. x-axis: normalized time (1000 time-points from the key-press of the initial note to the key-release of the final note). Each vertical dotted line indicates the moment of each key-press and the corresponding fingering (I: index, M: middle, R: ring, L: little).

Figure 4. Fundamental joint coordination patterns determined by PC analysis.

Group means of the weighting coefficients at the MCP (left), PIP (middle), and DIP (right) joints of the index (I), middle (M), ring (R) and little (L) fingers for the first 3 PCs (top 3 panels) and mean value that was subtracted from the inputted joint angular position prior to running the PC analysis (bottom panel). Each coefficient represents the degree to which the corresponding PC waveform constitutes the original motion at each joint and finger. Therefore a relation of the value and sign of the coefficients across joint and fingers at each PC describes the amount of movement covariation (or individuation). White and black bars indicate healthy controls and patients with FD, respectively. Error bars indicate one standard deviation within each of the groups. *p < 0.05, **p < 0.01. The hand in the figure was drawn by the corresponding author.

Figure 5. Rhythmic variability of keystrokes.

Group means of the inter-trial variability of key-striking and key-releasing movements at individual notes while healthy controls (white bars) and pianists with FD (black bars) were performing a sequence of eight successive keystrokes with the right hand. x-axis: note to be played (see the top-right small panel indicating the relationship between MIDI pitch number, note, and fingering; arrows indicate sequence of notes over time from top to bottom). (A) inter-keystroke interval, (B) inter-key-release interval, (C) finger-key contact duration, (D) overlap duration between two successive tones. Each of the variables was visually illustrated at the left-top small panel. Error bars indicate one standard deviation within each of the groups. *p < 0.05, **p < 0.01.