Pain evoked potentials (EPs) have been used in the last two decades as means of obtaining objective measures of pain, in clinical and experimental setups. The possibility that the pain EP wave contains elements of the endogenous P300 potential rather than being a neurophysiological correlate of pain has been raised by a number of authors, but the issue has not been resolved. In this study, two experiments were performed to study the effect of nonmodality-specific factors on the laser EP: (1) a stimulus attend as opposed to a stimulus-ignore condition and (2) counterbalanced oddball and task P300 stimulus presentations. The latter was to permit full examination of the separate and combined influences of each condition on the EP. Stimuli were given to the radial hand of 10 healthy volunteers using a CO2 laser. The positive component of the laser EP was affected by both manipulations relating to (1) attention (P = 0.0146) and (2) the frequency condition (P = 0.003) in the P300 paradigm. The task condition in the second paradigm did not affect the positive wave on its own, although its effect was visible in interaction with frequency (P = 0.033). In conclusion, although the presence of a somatic component in the laser EP cannot be rules out, we suggest that the laser EP contains a definite non-modality-specific P300 component, and is not a pure neurophysiological correlate of pain intensity.