After twenty there is marked change in the memory content. The male acquires more and the female less visual and auditory memories. The memories of the female are more logical, and topographical features increase. Memories of sickness and accidents to self decrease with the males and increase with the females, while in the case of both there is relative decline in the memories of sickness and accident to others. From all this it would appear that different memories culminate at different periods, and bear immediate relation to the whole mental life of the period. While perhaps some of the finer analyses of Colegrove may invite further confirmation, his main results given above are not only suggestive, but rendered very plausible by his evidence.
Statistics based upon replies to the question as to whether pleasant or unpleasant experiences were best remembered, show that the former increase at eleven, rise rapidly at fourteen, and culminate at eighteen for males, and that the curve of painful memories follows the same course, although for both there is a drop at fifteen. For females, the pleasant memories increase rapidly from eleven to thirteen, decline a little at fourteen, rise again at sixteen, and culminate at seventeen, and the painful memories follow nearly the same course, only with a slight drop at fifteen. Thus, up to twenty-two for males, there is a marked preponderance of pleasant over painful memories, although the two rise and fall together. After thirty, unpleasant memories are but little recalled. For the Indians and negroes in this census, unpleasant memories play a far more and often preponderating rôle suggesting persecution and sad experiences. Different elements of the total content of memory come to prominence at different ages. He also found that the best remembered years of life are sixteen to seventeen for males and fifteen for females, and that in general the adolescent period has more to do than any other in forming and furnishing the memory plexus, while the seventh and eighth year are most poorly remembered.
It is also known that many false memories insert themselves into the texture of remembered experiences. One dreams a friend is dead and thinks she is till she is met one day in the street; or dreams of a fire and inquires about it in the morning; dreams of a present and searches the house for it next day; delays breakfast for a friend, who arrived the night before in a dream, to come down to breakfast; a child hunts for a bushel of pennies dreamed of, etc. These phantoms falsify our memory most often, according to Dr. Colegrove, between sixteen and nineteen.
Mnemonic devices prompt children to change rings to keep appointments, tie knots in the handkerchief, put shoes on the dressing-table, hide garments, associate faces with hoods, names with acts, things, or qualities they suggest; visualize, connect figures, letters with colors, etc. From a scrutiny of the original material, which I was kindly allowed to make, this appears to rise rapidly at puberty.
[Footnote 1: See my Ideal School as Based on Child Study. Proceedings of the National Educational Association, 1901, pp. 470-490.]
[Footnote 2: Charles P.G. Scott: The Number of Words in the English and Other Languages. Princeton University Bulletin, May, 1902, vol. 13, pp. 106-111.]
[Footnote 3: The Teaching of English. Pedagogical Seminary, June, 1902, vol. 9, pp. 161-168.]
[Footnote 4: See my Some Aspects of the Early Sense of Self. American
Journal of Psychology, April, 1898, vol. 9, pp. 351-395.]
[Footnote 5: Sprachgeschichte und Sprachpsychologie, mit Rucksicht auf
B. Delbrück's "Grundfragen der Sprachforschung." Leipzig, W.
Engelmann, 1901]
[Footnote 6: Latin in the High School. By Edward Conradi. Pedagogical
Seminary, March, 1905, vol. 12, pp. 1-26.]