Learning effects during balance analysis on a modified posturomed-platform in healthy dogs

Abstract Background Posturographic balance assessments are increasingly used in veterinary medicine, yet potential learning effects during evaluation remain unstudied. This study investigated learning effects in static, slow-dynamic, and fast-dynamic posturography using a modified Posturomed platform in healthy dogs. Material and methods Healthy adult dogs (n = 20) were positioned longitudinally on a pressure sensitive modified balance platform (Posturomed-FDM-JS, Zebris, Isny, Germany). Five static, slow-dynamic and fast-dynamic posturographic trials were recorded (duration: 20s) and repeated three times over three weeks. Center of pressure (COP) parameters—COP-path-length (PL, mm), 95% COP-confidence-ellipse-area (CEA, mm2) and COP-average-velocity (AV, mm/sec) were analyzed over five steady-state 5-s intervals per trial. Data were analyzed using generalised linear or robust linear mixed-effects models with random effects on the individual dog; p-values were adjusted using the Tukey method for multiple comparisons. Results Under static conditions, none of the COP-parameters differed significantly across time points (all p-values > 0.448). Under slow-dynamic conditions, all COP-parameters decreased significantly between time points 1 and 2 (all p values < 0.0007) but remained stable thereafter (all p-values > 0.159). Under fast-dynamic conditions, all COP-parameters decreased significantly from time point to time point (all p-values < 0.034), except for CEA between time points 1 and 3 (p = 0.0759) and 2 and 3 (p = 0.999). Differences among trials occurred only at the first time point under dynamic conditions and were more pronounced under fast-dynamic conditions. Conclusions No learning effects were observed during static posturography in healthy adult dogs. However, training effects must be considered in both slow- and fast-dynamic posturography.