One form of confirmation bias is the tendency for people to ignore information that is inconsistent with their current beliefs. While confirmation bias is the subject of both analytical models and experiments in accounting and finance, its effect on market prices has not been studied due to limitations associated with traditional financial markets. In eleven real-money movie box office prediction markets, confirmation bias was induced in all traders via the explanation effect, i.e. a requirement to submit a box office forecast and an explanation prior to trading. When all traders are subject to confirmation bias, market prices do not accurately reflect new, value-relevant information. However, in comparable a set of seven real-money movie prediction markets that included both traders who have not been subject to explanation requirement and those who have, we find efficient incorporation of new information into market prices. This study extends our knowledge of the conditions under which individual trader biases affect market prices and provides potential insights into open questions about forecasting errors among financial analysts.