INTENSITY OF SENSATION. Every sensation has at least four attributes: Quality, clearness, duration and intensity. No one of these can be further defined than to say it is an aspect of immediate experience. By intensity of sensation, therefore, is meant that aspect of experience which, for example, we ascribe to tones as loud or faint, to colors as bright or chill, to pressures as heavy or light, to tastes and smells as strong or weak. It is important to distinguish between intensity of •a sensation and intensity of stimulus. The latter is physical, and is always measurable by some physical unit; the physical intensity of tones, for example, may be measured by amount of physical energy, weights by number of grams or ounces, lights by so much candle power, etc.
Intensity of sensation is partially conditioned upon intensity of stimulus. There are stimuli, however, which do not give rise to sensation; some tones are too faint to be heard. some weights too light to be felt, some solutions too weak to arouse taste, etc. We have then, as a first problem, to determine the intensity of stimulus which will just produce a sensation. Such a determination, to be valid, must be made under rigorously controlled conditions; the pos sibilities of error on the part of the observer are so numerous that the same type of methodi cal procedure is as necessary for its determina tion as for the just noticeable difference. (See DISCRIMINATION, SENSIBLE). The magnitude representing the stimulus which just arouses a sensation is known as the stimulus-limen, or RL. This may be defined as that stimulus which, in 50 per cent of a large number of observations, is reported as "present,* while in the remaining 50 per cent it is reported as 'not present.' It is a calculated value; and like the DL, it is an ideal, a most probable value.
We may, nevertheless, think of the RL as a point somewhere on the scale of stimulus-in tensity. The lower limit of this scale is zero intensity and it is, therefore, below the RL; the upper limit is infinity, since theoretically we may. gradually increase the energy of a tonal vibration, add candle power to candle power, pile gram upon gram without end. The in
tensity of sensation which corresponds to the RL is, on the other hand, the zero point of the scale of sensation-intensities; theoretically, the upper limit of this scale is again infinity; but practically it is subject to the capacity of the sense-organ.
We have seen that the unit of the stimulus scale is some physical unit like the gram or candle-power. What now is the unit of the sensation scale? The answer to this question is fundamental to the larger problem of mental measurement. Until the middle of the last century philosophers had said that mental meas urement was impossible because there was no unit of measurement in terms of sensation itself. Nevertheless, the astronomers had already di vided the stars into six magnitudes, and it has since been found that the difference in intensity between a star of the first and one of the second magnitude is approximately equal to that be tween stars of the second and third, third and fourth, fourth and fifth and fifth and sixth. Each magnitude is, then, a point on a scale, the points are eqni-distant as measured by dif ferences in intensity, and the difference between any two continguous points may be taken as the unit of the scale. We may say, therefore, that the difference between stars of the first and sixth magnitude is five times the difference be tween the first and second magnitudes. Such a unit, which we may call a supra-liminal dif. ference, is, of course, an arbitrary one, but it has proved to be adequate to the measurement of intensity in nearly if not all sense depart ments. A unit which gives less direct measure ments but which has a greater range of appli cability, because its magnitude is expressed in terms of the stimulus, is the just noticeable difference or the DL. We cannot assume off hand, to be sure, that all least differences are equal, that least steps at various parts of the scale are equal steps, but the experimental evi dence is, on the whole, favorable to this assump tion.