and simple enough to enable one to control the regard. It is difficult to verify this principle on the smaller figures which have finer patterns over which the eye involuntarily wanders from one point of regard to another.

The general law observed is that the point which first catches the eye or tends to hold its regard is perceived as a near part of the object. This fixes the view.

Why should the part which first catches the eye be judged near? It is a habit. Whenever we look at objects, economy bids us look first at the near side, other things being equal; this has resulted in the tendency of assuming that the first part seen is the near part of the object.

Why should the point upon which we maintain regard be judged near? It is a habit. Whenever we try to see a whole complex object there is a tendency to accommodate for a point near the center of the surface, and this is, as a rule, near us; the uniformity of this experience has resulted in the tendency of assuming that when we regard a fairly central point in a complex figure, this point is a near part of the object.

CHAPTER XIII

ATTENTION

For Two.*

IN reading this paragraph your attention moves from word to word, following close upon the movement of the point of regard, and you are probably aware of this movement of attention as an expression of personal activity, as an advancing wave in the tide of your feeling of interest, and as a complex of sensory impressions of bodily condition and processes of adjustment. These are the most direct ways in which we become aware of it; but attention is not an activity in itself, it is not essentially a feeling of interest, nor is it essentially awareness of the processes of bodily accommodation. Attention is the focus of consciousness as a whole, the state or form of concentration of the seeing, thinking, remembering, feeling, doing, etc., at a given moment. Attention is to consciousness as the point of regard is to the field of vision, the focal point.

1. Rhythm of Attention. When O is seated with eyes closed, in a quiet room, hold an open watch at such a distance from his ear that he can barely hear it. a signal let him direct his attention as steadily as

At

*This experiment must be performed in comparative quiet, which is usually best obtained in the evening.

possible upon the hearing of the sound for one minute and, with a pencil in hand, point upward when he hears the sound and downward when he does not hear it.* Following the second-hand on the watch, tap time quietly in the note-book, making a dotted line, thus, one dot per second; adopt two levels and trace on the upper when O points upward, and on the lower when he points downward, like this,

........

Preserve and number the curves as permanent records. Make a table showing the number of periods and the length of each period in seconds (a) for "Sound heard" and (b) for "Sound not heard." Make five trials.

This experiment demonstrates that attention fluctuates and is periodic. For reasons of biological economy, consciousness cannot remain focused upon a single unchanging object for more than a moment at a time; the focus, i.e., the attention, is intermittent.

A liminal stimulus was chosen because it is easier to observe the changes in that than in strong stimuli, although the principle of fluctuation applies to both, indeed perhaps to all objects of consciousness. And the sense of hearing was chosen because the stimulus can. be most satisfactorily controlled for that sense. In

*If the fluctuation does not occur, the watch is either too near or too far away; find such a distance that it will be heard about half of the time.

A common source of failure in a beginner is that he expects to hear the tick at one moment as well as at another and therefore fails to observe that he does not actually hear it but merely imagines that he hears it part of the time. An attentive observer may be able to record the missing of a single tick or two. If any wave in a curve is due to outside disturbance, discard that record and try it over again.

sight, touch, and smell the test is complicated by the rapid changes of adaptation and fatigue. The most satisfactory stimulus is a constant tone of medium pitch.

The waves of attention are not smooth. If we could get a detailed record of attention to a constant stimulus such as a sustained tone, a gray surface, or a uniform pressure, it would probably take a form something like the white surface in Fig. 27, which might represent the distribution of attention for about ten seconds schemat

*

ically. Both crests and troughs are rugged. In the case of a tone, for example, the result of a very faint tone might be represented at the level a-a', where the tone is heard very faintly and only for a relatively short period; the result for a stronger stimulus might be represented at the level b-b', where the tone is clearer and stronger and the periods of duration are relatively longer; while the result for a strong stimulus might be represented at the level c-c', which indicates that the observer has the impression of being aware of the tone

* For attention this is merely a hypothetical diagram; actually it is a photograph of the manometric flame for a vowel.

all the time but that there is a rhythmic fluctuation in its clearness and strength.*

Probably all mental activity-sensation, discrimination, feeling, memory, will, etc.-is rhythmic. The two waves in Fig. 27 are made up of smaller waves, and these in turn of still smaller ones. Measurements on long periods of attention show that there may be waves as long as an hour; smaller waves, reckoned in minutes, may be traced inside of these; and within the "minutewaves" we find the so-called "second-waves." Our conception of the flow of attention, or the fluctuation in the capacity for a certain conscious activity, is then this: There are infinitely short ripplets of attention, hardly perceptible; these form the surface of ripples, which in turn form the surface of wavelets; and these wavelets in turn form the surface of waves, and so on. Thus, all attention is rhythmic, and there is rhythm within rhythm from the infinitesimally short to the very long, even daily and annual periodicities.

The length of the attention-wave (second-wave) varies with numerous conditions, such as kind and strength of stimulus, practice, temperament, psychophysical condition, effort, etc. The shortest fluctuations are probably too rapid to be detected, and ordinarily a single crest of attention cannot be maintained for more than ten or twelve seconds. For conditions

*It is evident that the ratio of the length of the crests to the length of the troughs in the above experiment depended upon the strength of the stimulus adopted. When a stimulus is comparatively weak the troughs are long and the crests short; whereas the order is reversed when the stimulus is comparatively strong.