This can be generalized to include as many variables as you want.

In this case we have 3 variables:

Utility, Good A & Good B

Visualize utility measured on a third axis extending outward from this two dimensional graph. The utility function then becomes a “utility mountain.”

We slice this mountain at utility level 10 and project the contour of the slice back on our two dimensional graph.

In other words, we map utility in the same way that a hiking map of Colorado may show the level of the mountains using contour lines to show the elevation.