A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
The second derivative f''(x) is the derivative of f'(x), the third derivative f'''(x) is the derivative of f''(x), and so on. Physically, if f(t) is position, then f'(t) is velocity, f''(t) is acceleration, and f'''(t) is jerk. Higher-order derivatives reveal increasingly fine-grained information about how a function curves and changes. They are essential for concavity analysis, Taylor series, and differential equations.
Compute several derivatives of polynomial, trigonometric, and exponential functions to see patterns. Note that sin(x) cycles through sin, cos, -sin, -cos every four derivatives. Connect the second derivative to concavity and acceleration. Introduce notation: fn(x) or dn y/dxn.
Taking a derivative once answers the question "how fast is this changing?" Taking it again answers "how fast is that rate of change itself changing?" Each successive derivative zooms in on the behavior of the previous one. If f(x) is some smooth function, then f′(x) describes its slope at every point, f′′(x) describes how that slope is evolving, and f′′′(x) describes the evolution of the evolution of the slope. This is not mere abstraction: each order reveals genuinely new information that the previous order cannot tell you.
The physical interpretation is the clearest entry point. If f(t) is the position of a moving object at time t, then f′(t) is its velocity — the rate of change of position. But velocity itself changes, and the rate at which it changes is acceleration, f′′(t). This is what you feel pushing you back in your seat when a car speeds up: acceleration. There is even a name for f′′′(t): jerk — the rate of change of acceleration, which you feel as a lurch when a vehicle changes speed abruptly. Engineers designing roller coasters and elevator systems control jerk deliberately to prevent discomfort.
Applying the power rule repeatedly is straightforward for polynomials: if f(x) = x⁵, then f′(x) = 5x⁴, f′′(x) = 20x³, f′′′(x) = 60x², and so on until the function reaches zero. Trigonometric and exponential functions show more interesting patterns. The derivatives of sin(x) cycle with period 4: sin(x) → cos(x) → −sin(x) → −cos(x) → sin(x) → ⋯. The function ex is its own derivative at every order — this is the defining property that makes e special. These patterns become essential in Taylor series and differential equations, where higher derivatives encode the full local shape of a function.
Notation matters here because it is genuinely easy to confuse. The nth derivative fn(x) or dn y/dxn is a derivative applied n times; this is entirely different from f(x)n or (dy/dx)n, which are powers. The parentheses in fn exist precisely to signal "this is an order index, not an exponent." Similarly, d²y/dx² is not (dy/dx)²; the numerator exponent counts how many times you differentiate y, and the denominator exponent counts how many times x appears in the differential. Keeping this notation straight prevents a class of persistent errors when you encounter these expressions in concavity analysis, Taylor series, and ordinary differential equations.