The "fallacies of testimony" which depend on an adulteration of pure observation with inference and conjecture, as, for example, the inaccurate and wild statements of people respecting their experiences at spiritualist _seances_, while they illustrate the curious blending of both kinds of error, are probably much oftener illusions of memory than of perception.[125] Although in many cases we can account to ourselves for this confusion of fact and imagination, in other cases it is difficult to see any close relation between the fact remembered and the foreign element imported into it.

An idea of memory seems sometimes to lose its proper moorings, so to speak; to drift about helplessly among other ideas, and finally, by some chance, to hook itself on to one of these, as though it naturally belonged to it.

Anybody who has had an opportunity of carefully testing the truthfulness of his recollection of some remote event in early life will have found how oddly extraneous elements become incorporated into the memorial picture.

Incidents get put into wrong places, the wrong persons are introduced into a scene, and so on.

Here again we may illustrate the mnemonic illusion by a visual one.