the amount of time difference between buildup/release voltage of plugs gapped at .035 and plugs gapped at .060 would be measured in fractions of a microsecond, and are beyond the ecm's ability to even measure. more importantly, it would affect the timing by only about a 10,000th of a degree. since electronic ignitions use transistors to induce flyback voltage in a coil, the reaction speed is so high, that the time difference between the two gaps could not be the cause. if gap was to play a role here, i would say it's about 99% more likely that the smaller gap was not igniting the mixture occasionally, due to small spark frontal area, causing the o2 sensor to detect a "too rich" condition from the cycles where the combustion was either non existant, or weak. the ecm responds by leaning the mixture out, which now creates an overly lean condition in the cylinders (causes overheating and high nox) but since it's still not firing completely occasionally the mixture seems correct at the o2 sensor so the ecm thinks everything is ok. this scenerio seems far more plausable than saying that it takes longer for electricity to build up and travel an extra .015 inches, especially when you consider that it travels at the speed of light. even accounting for the difference it takes to get the cascade reaction & thus spark, we're talking a time differnce that your cars ecm could never measure, much less a timing light, or any other normal car diagnostic tool.