Background & Aims: Oropharyngeal dysphagia is an important disability that occurs after stroke; it contributes to aspiration pneumonia and death, and current modalities for rehabilitation of dysphagia have uncertain efficacy. We therefore examined the role of pharyngeal electrical stimulation (PES) in expediting human swallowing recovery after experimental (virtual) and actual (stroke) brain lesions. Methods: First, healthy subjects (n = 13) were given 1-Hz repetitive transcranial magnetic stimulation to induce a unilateral virtual lesion in pharyngeal motor cortex followed by active or sham (control) PES. Motor-evoked potentials and swallow accuracy were recorded before and after the lesion to assess PES response. Thereafter, 50 acute dysphagic stroke patients underwent either a dose-response study, to determine optimal parameters for PES (n = 22), or were assigned randomly to groups given either active or sham (control) PES (n = 28). The primary end point was the reduction of airway aspiration at 2 weeks postintervention. Results: In contrast to sham PES, active PES reversed the cortical suppression induced by the virtual lesion (F7,70 = 2.7; P = .015) and was associated with improvement in swallowing behavior (F3,42 = 5; P = .02). After stroke, 1 PES treatment each day (U = 8.0; P = .043) for 3 days (U = 10.0) produced improved airway protection compared with controls (P = .038). Active PES also reduced aspiration (U = 54.0; P = .049), improved feeding status (U = 58.0; P = .040), and resulted in a shorter time to hospital discharge (Mantel-Cox log-rank test, P = 0.038). Conclusions: This pilot study of PES confirms that it is a safe neurostimulation intervention that reverses swallowing disability after virtual lesion or stroke. © 2010 AGA Institute.