The present invention relates to methods of determining if a subject has an increased risk of suffering from memory impairment. The methods comprise determining the influx of polyunsaturated fatty acids (PUFAs) into the brain from plasma in a test subject and comparing this measured influx to a normal standard influx. An increase in influx of PUFAs from the plasma into the brain is indicative that the subject has an increased risk of suffering from memory impairment compared to a normal individual.