Exercise has been associated with increased arrhythmic tendency in athletes. Since reports have shown that horses have similar reporlarizing currents to humans, they may make a suitable electrophysiological model for human athletes. It is therefore important to determine whether the Thoroughbred horse system is amenable to standard electrophysiological analysis at incremental heart rates (HRs). To investigate this, we analysed the related adaptations in action potential (AP) restitution properties from Thoroughbred horses during field exercise. Electrocardiographs (ECGs) from seven Thorougbred horses were recorded non-invasively as part of routine clinical workup. The RR interval, QRS duration, QT and TQ intervals provided indications of basic cycle length (BCL), conduction velocity (CV), action potential duration (APD), and diastolic interval (DI) respectively. Further, the presence of QT interval alternans was assessed. From these variables, indices of active (λ = (QT interval)/(QRS duration)) and resting (λ0 = (TQ interval)/(QRS duration)) AP wavelengths were calculated. Both restitution plots of QT interval against TQ interval, and of λ against λ0, followed the function [endif]–> (Matthews et al., 2013). Data points were obtained over a range of BCL intervals, corresponding to extrapolated heart rates (eHR) (eHR (beats per minute (bpm)) = (1/(BCL in seconds) × 60)) of between 24 and 169 bpm. Assessment of QT interval alternans (64 episodes across seven horses) showed that episodes were transient and not restricted to high HR, making it unlikely to be associated with major arrhythmias. On analysis of QT interval and λ restitution plots, critical eHR values (eHRcrit; the value of eHR when restitution plot attains unity slope (gradient (τ) = 1)) were 127.7±11.3 and 117.4±5.6 bpm respectively (mean± SEM; ANOVA: p>0.05; non-significant). Correspondingly, the permissible eHRs (permissible eHR = eHRcrit- average minimum her observed (eHRmin)) were 94.7±10.8 and 84.4±4.9 bpm respectively (ANOVA: p>0.05; non-significant). This accommodation of the increase in eHR was achieved by shortening QT interval to 60% or λ to 66% of their maximum value. The study suggests a basis for the range between eHRmin and eHRcrit in the absence of sustained alternans. Restitution plots of QT interval and λ assumed a plateau at the lower eHRs and declined only at the highest eHRs. This accommodation of QT interval and λ at high eHRs in turn mitigates the reduction in TQ interval and λ0 with increasing eHR. Our findings have supported this hypothesis. This enables an enhanced time for the tissue to recover from refractoriness following excitation, explaining why horses are capable of showing high HRs without compromising their electrophysiological stability. This study supports the use of Thoroughbred horses as a model for human athletes.