In the latter case we purposely abstract from the perceiving subject, and under circumstances, especially when favored by the regularity of the phenomena, as in astronomy, mere observation may succeed in determining with adequate certainty the objective components of the processes. Psychology, on the contrary, is debarred from this abstraction by its fundamental principles, and the conditions for chance observation can be suitable only when the same objective components of immediate experience are frequently repeated in connection with the same subjective states. It is hardly to be expected, in view of the great complexity of psychical processes, that this will ever be the case. The coincidence is especially improbable since the very intention to observe, which is a necessary condition of all observation, modifies essentially the rise and progress of psychical processes. Observation of nature is not disturbed by this intention on the part of the observer, because here we purposely abstract from the state of the subject. The chief problem of psychology, however, is the exact observation of the rise and progress of subjective processes, and it can be readily seen that under such circumstances the intention to observe either essentially modifies the facts to be observed, or completely suppresses them. On the other hand, psychology, by the very way in which psychical processes originate, is led, just as physics and physiology are, to employ the experimental mode of procedure. A sensation arises in us under the most favorable conditions for observation when it is caused by an external sense-stimulus, as, for example, a tone-sensation from an external tone-vibration, or a light-sensation from an external light-impression. The idea of an object is always caused originally by the more or less complicated cooperation of external sense-stimuli.
If we wish to study the way in which an idea is formed, we can choose no other method than that of imitating this natural process. In doing this, we have at the same time the great advantage of being able to modify the idea itself by changing at will the combination of the impressions that cooperate to form it, and of thus learning what influence each single condition exercises on the product. Memory-images, it is true, cannot be directly aroused through external sense impressions, but follow them after a longer or shorter interval. Still, it is obvious that their attributes, and especially their relation to the primary ideas through direct impressions, can be most accurately be learned, not by waiting for their chance arrival, but by using such memory-ideas as may be aroused in a systematic, experimental way, through immediately preceding impressions. The same is true of feelings and volitions; they will be presented in the form best adapted to exact investigation when those impressions are purposely produced which experience has shown to be regularly connected with affective and volitional reactions. There is, then, no fundamental psychical process to which experimental methods can not be applied, and therefore none in whose investigation they are not logically required.