Points of inflection occur where a curve changes concavity, switching from bending upward to bending downward or vice versa. In calculus terms, this happens where the second derivative changes sign. Finding these points requires checking where the second derivative equals zero or is undefined, then confirming the sign actually flips.
What Concavity Means
A function is concave up when its graph curves like a bowl, opening upward. Visually, the slope is getting steeper (or less negative) as you move to the right. The second derivative is positive in these regions. A function is concave down when it curves like an upside-down bowl, and the second derivative is negative.
An inflection point is the exact spot where the curve transitions between these two shapes. Think of driving on a road that curves left for a while, then begins curving right. The moment the road straightens out before bending the other way is the inflection point.
The Second Derivative Must Change Sign
The core requirement is simple: the second derivative, f”(x), must switch from positive to negative or from negative to positive at that point. This sign change is what defines an inflection point, not merely the second derivative hitting zero.
For the sign to change, the second derivative must either equal zero or be undefined at the transition. So the process starts by finding all x-values where f”(x) = 0 or where f”(x) doesn’t exist. These are your candidates. But candidates aren’t automatically inflection points. You still need to verify that the sign of f”(x) is different on each side of the candidate.
This distinction trips up a lot of students. The classic counterexample is f(x) = x⁴. The second derivative is 12x², which equals zero at x = 0. But 12x² is positive on both sides of zero, so the concavity never changes. The function is concave up everywhere, and x = 0 is not an inflection point despite the second derivative being zero there.
How to Find Inflection Points Step by Step
Start by computing the second derivative of your function. Then set f”(x) = 0 and solve. Also identify any x-values where f”(x) is undefined but the original function f(x) is defined. These are all your candidate points.
Next, build a sign chart. Pick test values on each side of every candidate and plug them into f”(x). If the sign flips from positive to negative (or negative to positive) across a candidate, that candidate is a confirmed inflection point. If the sign stays the same on both sides, discard it.
One detail worth noting: the original function must be defined at the candidate point. If f(x) itself doesn’t exist at some x-value, that point can’t be an inflection point because it’s not actually on the graph. However, the second derivative can be undefined at an inflection point, as long as the original function is defined there. A common example is f(x) = x^(1/3), where f”(x) is undefined at x = 0, but the function itself is defined and the concavity changes at that point.
The Connection to the First Derivative
There’s a useful relationship between inflection points and the first derivative. The sign of f'(x) tells you whether f(x) is increasing or decreasing. The sign of f”(x) tells you how the function is increasing or decreasing. When f”(x) is positive, the function is increasing at an accelerating rate (or decreasing at a slowing rate). When f”(x) is negative, the opposite is true.
Because an inflection point is where f”(x) changes sign, it corresponds to a local maximum or minimum of f'(x). In other words, the slope of the original function reaches a peak or a valley at an inflection point. The curve is getting steeper and steeper, hits the inflection point, and then starts getting less steep, or vice versa.
A Worked Example
Suppose f(x) = x³ − 6x² + 9x + 1. The first derivative is f'(x) = 3x² − 12x + 9, and the second derivative is f”(x) = 6x − 12. Setting the second derivative to zero gives 6x − 12 = 0, so x = 2.
Now check the sign on each side. At x = 1, f”(1) = 6(1) − 12 = −6, which is negative (concave down). At x = 3, f”(3) = 6(3) − 12 = 6, which is positive (concave up). The sign changes from negative to positive, so x = 2 is a confirmed inflection point. Plugging back in, f(2) = 8 − 24 + 18 + 1 = 3, so the inflection point is at (2, 3).
Inflection Points in Real-World Models
Outside the classroom, inflection points show up wherever growth changes its character. In economics, the point of diminishing returns on a production curve is an inflection point. Before that point, each additional unit of input (say, hiring one more worker) produces a bigger jump in output than the last. After it, each additional unit produces a smaller jump. The total output is still increasing, but the rate of increase has started to slow down.
For example, consider a production function R = −2x³ + 24x² + 50. The second derivative is R” = −12x + 48, which equals zero at x = 4. Before x = 4, adding resources accelerates output. After x = 4, adding the same resources yields progressively smaller gains. The inflection point at x = 4 marks the most efficient level of production, with a total output of 306 units.
Population growth curves follow the same logic. In a sigmoid (S-shaped) growth curve, the inflection point is where growth is fastest. Before it, the population is accelerating. After it, the population is still growing but decelerating as it approaches a carrying capacity. Epidemic curves, technology adoption rates, and learning curves all share this same inflection point structure, making it one of the most broadly useful concepts in applied mathematics.