I may, perhaps, illustrate these two classes of illusion by the simile of an interpreter poring over an old manuscript.  The first would be due to some peculiarity in the document misleading his judgment, the second to some caprice or preconceived notion in the interpreter’s mind.

It is not difficult to define conjecturally the physiological conditions of these two large classes of illusion.  On the physical side, an illusion of sense, like a just perception, is the result of a fusion of the nervous process answering to a sensation with a nervous process answering to a mental image.  In the case of passive illusions, this fusion may be said to take place in consequence of some point of connection between the two.  The existence of such a connection appears to be involved in the very fact of suggestion, and may be said to be the organic result of frequent conjunctions of the two parts of the nervous operation in our past history.  In the case of active illusions, however, which spring rather from the independent energy of a particular mode of the imagination, this point of organic connection is not the only or even the main thing.  In many cases, as we shall see, there is only a faint shade of resemblance between the present impression and the mental image with which it is overlaid.  The illusions dependent on vivid, expectation thus answer much less to an objective conjunction of past experiences than to a capricious subjective conjunction of mental images.  Here, then, the fusion of nervous processes must have another cause.  And it is not difficult to assign such a cause.  The antecedent activity of imagination doubtless involves as its organic result a powerful temporary disposition in the nervous structures concerned to go on acting.  In other words, they remain in a state of sub-excitation, which can be raised to full excitation by a slight additional force.  The more powerful this disposition in the centres involved in the act of imagination, the less the additional force of external stimulus required to excite them to full activity.

Considering the first division, passive illusions, a little further, we shall see that they may be broken up into two sub-classes, according to the causes of the errors.  In a general way we assume that the impression always answers to some quality of the object which is perceived, and varies with this; that, for example, our sensation of colour invariably represents the quality of external colour which we attribute to the object.  Or, to express it physically, we assume that the external force acting on the sense-organ invariably produces the same effect, and that the effect always varies with the external cause.  But this assumption, though true in the main, is not perfectly correct.  It supposes that the organic conditions are constant, and that the organic process faithfully reflects the external operation.  Neither of these suppositions is strictly true.  Although in general we may abstract from the organism and view the relation between the external fact and the mental impression as direct, we cannot always do so.