Distinct modulation of event-related potentials during motor preparation in patients with motor conversion disorder

Abstract

Objective: Conversion paresis patients and healthy people feigning weakness both exhibit weak voluntary movement without detectable neuropathology. Uniquely, conversion patients lack a sense of conscious awareness of the origin of their impairment. We investigated whether conversion paresis patients show distinct electroencephalographic (EEG) markers associated with their unconscious movement deficits.

Methods: Six unilateral upper limb conversion paresis patients, 12 feigning participants asked to mimic weakness and 12 control participants performed a precued reaction time task, requiring movements of either hand, depending on precue information. Performance measures (force, reaction and movement time), and event-related EEG potentials (ERP) were compared, between groups and across hands or hemisphere, using linear mixed models.

Results: Feigners generated the same inter-hand difference in reaction and movement time as expressed by patients, even though no specific targets were set nor feedback given on these measures. We found novel ERP signatures specific to patients. When the symptomatic hand was precued, the P3 ERP component accompanying the precue was dramatically larger in patients than in feigning participants. Additionally, in patients the earlier N1 ERP component was diminished when the precue signalled either the symptomatic or asymptomatic hand.

Conclusions: These results are consistent with previous suggestions that lack of awareness of the origin of their symptoms in conversion disorder patients may result from suppression of brain activity normally related to self-agency. In patients the diminished N1 to all precues is consistent with a generalised reduction in cognitive processing of movement-related precues. The P3 enhancement in patients is unlikely to simply reflect changes required for generation of impaired movements, because it was not seen in feigners showing the same behavioural deficits. Rather, this P3 enhancement in patients may represent a neural biomarker of unconscious processes, including additional emotional loading, related to active suppression of brain circuits involved in the attribution of self-agency.

Figure 1. Experimental task and behavioral results.

(A) Temporal sequence for each trial. Circles in black rectangle represent response key positions under the middle and index fingers of each hand. In this example, the blue precue symbol signals preparation of the left hand-middle finger. After a foreperiod of 1500 ms, the same blue symbol is illuminated as the imperative stimulus. (B) Mean reaction time (+ SEM) and (C) mean movement time (+ SEM) for the symptomatic (left) hand (black bars) and the asymptomatic (right) hand (grey bars) from patients (n = 6), feigners (n = 12), and controls (n = 12). *P<.05, **P<.01.

Figure 2. Grand mean EEG waveforms.

(A) Symptomatic (left) hand, recording from contralateral occipital cortex. Top plot shows average across entire trial duration, dashed lines at t = −2.0 s and t = 0 s indicate precue and imperative stimulus onsets, grey horizontal bar shows precue duration. Inset panels zoom on precue onset (left; visual ERP) and before stimulus onset (right; terminal CNV). (B) Asymptomatic (right) hand, recording from contralateral occipital cortex. (C) As for A, recording from motor cortex contralateral to the symptomatic hand. (D) As for B, recording from motor cortex contralateral to the asymptomatic hand. Negative upwards in all plots.

Figure 3. Quantification of ERP measures.

(A–B) Mean (+ SEM) occipital N1 amplitude for symptomatic and asymptomatic hand precues respectively. Black bars, left hemisphere; grey bars right hemisphere. (C–D) P3 amplitudes at central electrodes. (E–F) P3 amplitudes at occipital electrodes. (G–H) CNV amplitudes at central electrodes. Negative upwards in all graphs. *P<.05, **P<.01.