Previous research with memory questionnaires has been interpreted as showing that people do not recall their memory failures accurately. The present research consisted of three investigations that examined whether participants' reports of memory failures were corroborated by their spouses. These investigations used a greater variety of statistical and survey methods than were used in previous investigations. The results demonstrated that people are able to recall memory failures more accurately than previous research indicated. These results have important methodological and theoretical implications for basic and applied assumptions about what people know about their memory failures. Copyright © 2005 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]