Thus, if older adults relative to younger adults show some advantage in illnesses alone, we might surmise that experience in this one domain is driving the result, whereas if older adults show a generalized advantage or disadvan- tage across domains, then we would conclude that the effect is not specific to the domain of illnesses. Many investigators consider that a specifically cultural domain assumes shared knowledge, and at the absolute minimum, at least two indi- viduals must have mentioned an item to warrant its inclusion in the group list. On the other hand, where domains are clearly shared by young and old, it is an open ques- tion whether age has discernible effects. Our elicitation procedure involved simply asking each participant to list as many items as he or she could for each category. Another method for boosting responses is the use of more specific prompts. Obviously, the vocabulary tasks in these studies differ from free listing, but they do offer some clues about age differences. Note that small variance in young and old MMSE scores yields a statistically significant but substantively meaningless difference between the groups. The younger group included students seventeen , sales personnel three , office workers six , business owners or employees three , and a govern- ment employee and a reporter. This is consistent with the general finding in cognitive aging that older adults are disadvantaged relative to younger adults on more effortful production tasks than on more automatic recognition tasks Park There is a great deal of research in this area, but the conclusions are critically dependent on task type and cohort effects. Does the age of the informant make a difference in free list production? The cost may be that the resulting list of items will not be as comprehensive as when older adults are included. Again, semantic fluency differs from free listing in that the former is timed but the latter is not. Thus, items mentioned by only one individual may belong to a domain and be widely recognized as such but may appear idiosyncratic in the analysis. It should be noted, however, that a variety of methods exist to boost free listing output. That is, it is possible that older adults will produce fewer or equivalent numbers of items as young adults in their individual lists but a greater number of unique items at the level of the group.