Neural constraints on information processing have emerged as a possible general explanation for why a majority of plant-feeding insects are relative specialists. According to the hypothesis, acquiring and processing information necessary to make fast and accurate oviposition decisions will carry a cost. As plants constitute a vast but diverse resource, the cost can be expected to increase with increasing host range. The cost can be paid in two currencies: time or accuracy. Both types of costs have been demonstrated, using a variety of taxa including butterflies. However, all studies have measured either one of the two currencies. Hence, there is the possibility that a decrease in one of the currencies can be compensated by an increase in the other, in which case the net outcome may not be a cost at all. Poor oviposition decisions could then be counterbalanced by shorter decision times, which could lead to higher realized fecundity. Using two strains of butterflies (Polygonia c-album) with different degrees of specificity, I test the hypothesis that the previously reported higher error rates in the more polyphagous butterflies are compensated by shorter decision times at oviposition. Post-alighting decision times were measured while the females evaluated a preferred plant (Urtica dioica) for oviposition. Contrary to the expectations, the polyphagous butterflies actually had longer decision times than the specialized butterflies, giving no support for the quality vs. quantity hypothesis. Instead, the results suggests that the polyphagous butterflies do pay a real cost for their wide host range and thus gives increased support for the information processing hypothesis as a general explanation for the widespread host specificity among plant-feeding insects.