Objectives. The aim of this study was to investigate the pathophysiology of persistent geotropic direction-changing positional nystagmus (pGeo DCPN) by analyzing caloric responses and the clinical course in patients with pGeo DCPN and other variants of lateral semicircular canal benign paroxysmal positional vertigo (LSCC BPPV).Methods. In this prospective case-control study, 101 patients diagnosed with pGeo DCPN (pGeo group, n=34), persistent apogeotropic (pApo) DCPN (pApo group, n=40), or transient geotropic (tGeo) DCPN (tGeo group, n=27) involving the LSCC were enrolled. The video head impulse test (vHIT) and caloric test were performed at diagnosis. If one or both tests were abnormal, follow-up testing was conducted after nystagmus resolution. Differences in test results and clinical courses among the LSCC BPPV types were analyzed and compared.Results. The mean disease duration was significantly longer in patients with pGeo (23.0 days) than those with pApo (8.0 days) and tGeo (9.0 days) (P<0.005). All patients demonstrated normal vHIT gain. The pGeo group showed the highest canal paresis (CP) value (vs. pApo: P=0.002; vs. tGeo: <i>P<0.001) and a higher frequency of abnormal CP (61.8%) than the pApo (22.5%) and tGeo (11.1%) groups. CP was predominantly ipsilesional (85.7%) in the pGeo group. Follow-up caloric tests indicated decreased CP in 72.7% of pGeo patients, with normalization in 36.4% following the disappearance of nystagmus.Conclusion. The clinical course, caloric tests, and vHIT results indicate distinct pathophysiological mechanisms for different LSCC BPPV types. Cupular deflection due to buoyancy in pGeo DCPN plausibly leads to CP while preserving vHIT integrity.