Blood and urine samples were collected before and after each dose and at eight weeks after DHA supplementation stopped.

Researchers then examined these samples for biochemical markers indicating the effects of each dose on the volunteers.

'Now that we have a very good idea about how much DHA is just right, the next step is to try it out in an expanded clinical trial that involves many more people,' said Gerald Weissmann, editor-in-chief of The FASEB Journal, which is slated to publish the report.

'Until then, I'll stick with tasty foods that contain DHA, like fish, rather than getting a quick fatty-acid fix at the local vitamin store,' Weissmann said.

FASEB is the Federation of American Societies for Experimental Biology.