Academic journal article Human Factors

Isoperformance Curves in Applied Psychology

Academic journal article Human Factors

Isoperformance Curves in Applied Psychology

Article excerpt


Isoperformance is based on an old idea. The usual relationship between two variables involves a determinant and an effect or, somewhat more generally, an independent and a dependent variable. Typically the determinant or independent variable is presented on the abscissa and the effect or dependent variable on the ordinate. Thus response is plotted as a function of dose, or income as a function of age or schooling.

Occasionally one sees a different relationship between two variables - namely, the relationship between two determinants that together produce a constant effect. In such a case both abscissa and ordinate are determinants. The plotted points represent equivalent combinations of the two determinants - equivalent in the sense that any one such combination produces the same specified effect as any other. In economics, relationships of this second type are usually called indifference curves. In sensory psychology and physiology, they are often called isofrequency, isochronal, or isoelectric curves or contours, or something else with the prefix iso-.

Psychophysics is the only psychological science in which plots of constant effect are frequently found. Figure 1 presents a familiar example, the scotopic luminosity curve. The abscissa is wavelength in [[micro]meter] and the ordinate, luminosity, is the reciprocal of the radiant flux at a given wavelength sufficient to produce a constant visual effect (e.g., threshold or matching brightness), relative to the maximum at 510 [[micro]meter]. There is no suggestion here that wavelength is a determinant of radiant flux. Rather, the point is that the amount of radiant flux necessary to produce the same effect on the eye varies with wavelength.

Isoperformance is not merely a formal device - that much it shares with isoappearance. It also carries a burden of content and context that contrasts sharply with that carried by the psychophysical curves, in two respects especially. First, isoperformance always involves an objective that the user is trying to achieve, a goal, an applied purpose that he or she is trying to realize. Isoappearance involves nothing of the sort. There is, after all, no difficulty in seeing a patch of light or matching brightnesses. A luminosity curve is simply an effective way to organize information.

Second, although isoappearance curves are, in fact, trade-off functions, psychophysicists do not think of them that way. A luminosity curve is primarily a way of describing the sensitivity of the eye to light of different wavelengths. The trade-off between radiant flux and wavelength is incidental. It has no interest or importance of its own.

In isoperformance, however, trade-off functions are central, mainly because of what is being traded off. It costs money to train workers or military personnel. It also costs money to recruit high-aptitude people or to engineer equipment to be usable by low-aptitude people. Trading off is an economic idea, and it is important outside economics only when the stakes involved, when the variations traded off, have heavy economic or human implications.

Isoperformance is a cost-effectiveness method. The usual approach to cost-effectiveness is to maximize effectiveness subject to constraints such as cost, safety, and feasibility. The same elements reappear in isoperformance but in altered roles. Effectiveness is no longer to be maximized. Instead, a particular level of effectiveness is to be singled out in advance and stated as an objective. Maximization is replaced by the derivation of isoperformance curves. The constraints, too, are still there, but they no longer serve as boundaries on a maximization process. In isoperformance, considerations such as cost, safety, feasibility, and (usually, to some degree) arbitrary decisions serve as secondary criteria for the selection of a single solution out of the many defined by an isoperformance curve. Altogether, isoperformance is less goal-oriented than the usual approach and more involved with means - that is, with the interplay of determinants both with one another and with what we have just called secondary criteria. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.