But what, it may be asked, are these false and illegitimate sources of mnemonic images, these unauthorized mints which issue a spurious mental coinage, and so confuse the genuine currency?  They consist of two regions of our internal mental life, which most closely resemble the actual perception of real things in vividness and force, namely, dream-consciousness and waking imagination.  Each of these may introduce into the mind vivid images which afterwards tend, under certain circumstances, to assume the guise of recollections of actual events.

That our dream-experience may now and again lead us into illusory recollection has already been hinted.  And it is easy to understand why this is so.  When dreaming we have, as we have seen, a mental experience which closely approximates in intensity and reality to that of waking perception.  Consequently, dreams may leave behind them, for a time, vivid images which simulate the appearance of real images of memory.  Most of us, perhaps, have felt this after-effect of dreaming on our waking thoughts.  It is sometimes very hard to shake off the impression left by a vivid dream, as, for example, that a dead friend has returned to life.  During the day that follows the dream, we have at intermittent moments something like an assurance that we have seen our lost friend; and though we immediately correct the impression by reflecting that we are recalling but a dream, it tends to revive within us with a strange pertinacity.

In addition to this proximate effect of a dream in disturbing the normal process of recollection, there is reason to suppose that dreams may exert a more remote effect on our memories.  So widely different in its form is our dreaming from our waking experience, that our dreams are rarely recalled as wholes with perfect distinctness.  They revive in us only as disjointed fragments, and only for brief moments when some accidental resemblance in the present happens to stir the latent trace they have left on our minds.  We get sudden flashes out of our dream-world, and the process is too rapid, too incomplete for us to identify the region whence the flashes come.

It is highly probable that our dreams are, to a large extent, answerable for the sense of familiarity that we sometimes experience in visiting a new locality or in seeing a new face.  If, as we have found some of the best authorities saying, we are, when asleep, always dreaming more or less distinctly, and if, as we know, dreaming is a continual process of transformation of our waking impressions in new combinations, it is not surprising that our dreams should sometimes take the form of forecasts of our waking life, and that consequently objects and scenes of this life never before seen should now and again wear a familiar look.