Experimental pain models inducing hyperalgesia, i.e. an increased sensitivity to noxious stimuli often present in clinical pain, are important tools for studying antinociceptive drug profiles. The correct interpretation of results obtained in these models necessitates their mechanistic understanding. This study evaluated the freeze lesion, an experimental model of hyperalgesia, in humans. Twelve healthy subjects were tested with mechanical (brush, punctuated and blunt) and electrical (5, 250, and 2000 Hz sine wave current) stimuli before and after freezing the skin, and during a computer-controlled infusion of the mu-opioid agonist remifentanil targeting five different plasma concentrations between 0 and 6 ng/ml in a two-staged, single occasion, randomized, and double blind study design. Pharmacodynamic modeling techniques were used to describe the effect of freezing and drug administration on the mechanical and electrical pain thresholds. Freezing the skin resulted in hyperalgesia to blunt and punctuated stimuli and lowered the respective pain threshold by 29 and 73%. Hyperalgesia to brushing or electrical stimuli was not detected. Remifentanil attenuated hyperalgesia to blunt stimuli about twice as potently as hyperalgesia to punctuated stimuli, as indicated by a significantly steeper linear relationship between the remifentanil plasma concentration and the increase of the pain threshold to blunt stimuli. Remifentanil attenuated electrical pain with greater potency for low frequency stimulation. The potency difference of remifentanil suggests that different neuronal mechanisms mediate hyperalgesia to blunt and punctuated stimulation. Absence of brush-evoked and electrical hyperalgesia is compatible with the view that mechanical hyperalgesia to blunt and punctuated stimulation of the freeze lesion is predominantly caused by a peripheral mechanism.