The sensation of cooling appears to depend largely on the transient receptor potential channel TRPM8. However, the detection of noxious cold is still controversial. Many targets have been proposed for the transduction of cold into an electrical signal. So far, there is no experimental human model that allows the application of potential antagonists at the same site as the cold. This is further complicated by the fact that the site of cold transduction is unclear.We have developed a human pain model based on intradermal injection of cooled extracellular solution into the volar forearm. This is based on a largely linear decrease in the temperature of the solution down to 3°C, allowing assessment of a threshold, followed by a cold plateau to assess further time dependence. Volunteers periodically rated the perceived pain during the injection using a numerical rating scale. As a negative control, the same solution was applied without cooling. Lidocaine was added as a positive control.Cold pain thresholds showed considerable variation, as described before. It is therefore important that the established protocol was suitable for inducing cold pain in all subjects. No side effects were observed. The addition of lidocaine reduced cold-induced pain, providing proof of concept for pharmacological inhibition.As demonstrated for acid-induced pain, the injection-based pain model allows a noxious cold stimulus to be reliably co-administered with a compound of interest at the same site. This can serve as a model for testing antagonists to elucidate the molecular entities contributing to cold pain.