AP Calculus BC Unit 3 Notes: Derivatives Involving Inverses

Differentiating Inverse Functions

What an inverse function is (and what it is not)

An inverse function reverses the action of a function. If a function $f$ takes an input $x$ and produces an output $y$, then its inverse $f^{-1}$ takes that output $y$ back to the original input $x$.

Formally, when $f$ is one-to-one on a domain (so each output comes from exactly one input), the inverse function $f^{-1}$ is defined by the relationship:

$f(f^{-1}(x)) = x$

and also

$f^{-1}(f(x)) = x$

These equations are the “undoing” idea written precisely.

A common confusion is the notation: $f^{-1}(x)$ does **not** mean $\frac{1}{f(x)}$. It means “the inverse function of $f$ evaluated at $x$.”

Notation reference (very testable)

Why inverse derivatives matter

In calculus, you often know something about $f$ (a formula, a table, a graph, or values of $f'$), but you need information about $f^{-1}$. Inverse derivatives let you compute slopes of inverse functions without explicitly solving for $f^{-1}$.

This is powerful on AP problems because:

• You might be given a table of $f$ and $f'$ values and asked for $\left(f^{-1}\right)'$ at a point.
• You might be able to avoid messy algebra by using the inverse derivative relationship.
• It connects naturally to chain rule thinking: compositions like $f(f^{-1}(x))$ simplify in a way that creates a derivative formula.

The key idea: inverse functions swap inputs and outputs

If $f(a) = b$, then $f^{-1}(b) = a$. Graphically, $f$ and $f^{-1}$ are reflections across the line $y = x$. That reflection also swaps the roles of “run” and “rise,” which hints at why slopes are reciprocals (when they exist).

The derivative of an inverse function (main formula)

To differentiate an inverse function, start from the identity:

$f(f^{-1}(x)) = x$

Differentiate both sides with respect to $x$. The left side requires the chain rule:

$f'(f^{-1}(x))\left(f^{-1}\right)'(x) = 1$

Solve for $\left(f^{-1}\right)'(x)$:

$\left(f^{-1}\right)'(x) = \frac{1}{f'(f^{-1}(x))}$

This formula is the general inverse-derivative relationship.

A point-specific version (often easiest on AP)

If $f$ is differentiable at $a$ and $f'(a) \neq 0$, and if $f(a) = b$, then:

$\left(f^{-1}\right)'(b) = \frac{1}{f'(a)}$

This version is extremely practical because it turns “derivative of the inverse at $b$” into “reciprocal of derivative of the original at the corresponding input $a$.”

When the formula applies (and when it fails)

You need two main conditions at the relevant point:

1. $f$ must be **one-to-one** on an interval so that $f^{-1}$ is actually a function.
2. $f'(a) \neq 0$ at the matching point. If $f'(a) = 0$, the inverse would have an undefined or infinite slope there (a vertical tangent), so a finite derivative for $f^{-1}$ cannot be computed via the reciprocal.

Also, if $f$ has a cusp or corner (not differentiable) at $a$, then $f^{-1}$ may fail to be differentiable at $b$ as well.

Example 1: Using the point-specific inverse derivative rule

Suppose you are told:

• $f(2) = 5$
• $f'(2) = 3$

Find $\left(f^{-1}\right)'(5)$.

Reasoning: Since $f(2)=5$, that means $f^{-1}(5)=2$. The inverse derivative at output $5$ uses the reciprocal of the original derivative at input $2$.

$\left(f^{-1}\right)'(5) = \frac{1}{f'(2)}$

$\left(f^{-1}\right)'(5) = \frac{1}{3}$

What could go wrong: A very common mistake is to write $\left(f^{-1}\right)'(2)=\frac{1}{f'(2)}$. The input to the inverse derivative must be the original output (here, $5$), not the original input.

Example 2: Using the general formula without solving explicitly

Let:

$f(x) = x^3 + x + 1$

Compute $\left(f^{-1}\right)'(1)$.

Step 1: Find the matching input. You need $a$ such that $f(a)=1$.

$a^3 + a + 1 = 1$

$a^3 + a = 0$

$a(a^2 + 1) = 0$

So $a = 0$ is the real solution, and $f(0)=1$.

Step 2: Differentiate $f$ and evaluate at $a$.

$f'(x) = 3x^2 + 1$

$f'(0) = 1$

Step 3: Apply the inverse derivative rule.

$\left(f^{-1}\right)'(1) = \frac{1}{f'(0)}$

$\left(f^{-1}\right)'(1) = 1$

What could go wrong: Students sometimes try to solve for $f^{-1}(x)$ explicitly (which can be impossible in closed form for many functions). You usually do not need an explicit inverse.

Example 3: A table-based AP-style question

You are given values:

Find $\left(f^{-1}\right)'(7)$.

Because $f(3)=7$, you know $f^{-1}(7)=3$, so:

$\left(f^{-1}\right)'(7) = \frac{1}{f'(3)}$

$\left(f^{-1}\right)'(7) = \frac{1}{-5}$

$\left(f^{-1}\right)'(7) = -\frac{1}{5}$

Notice the negative sign: if $f$ is decreasing at $x=3$, then $f^{-1}$ is also decreasing at $x=7$.

Exam Focus

• Typical question patterns:
  • Given $f(a)=b$ and $f'(a)$, find $\left(f^{-1}\right)'(b)$.
  • Given a table or graph of $f$ and slopes, find the slope of $f^{-1}$ at a specified input.
  • Use $\left(f^{-1}\right)'(x) = \frac{1}{f'(f^{-1}(x))}$ as part of a larger chain-rule computation.
• Common mistakes:
  • Mixing up inputs: using $\left(f^{-1}\right)'(a)$ instead of $\left(f^{-1}\right)'(b)$ where $b=f(a)$.
  • Treating $f^{-1}$ as a reciprocal: confusing $f^{-1}(x)$ with $\frac{1}{f(x)}$.
  • Forgetting the condition $f'(a)\neq 0$ and concluding a finite derivative when the inverse actually has a vertical tangent.

Differentiating Inverse Trigonometric Functions

What inverse trig functions mean (conceptually)

Inverse trigonometric functions answer questions like: “Which angle has sine equal to this number?” For example, $\arcsin(x)$ is the angle (in radians on AP) whose sine is $x$.

However, trig functions like sine are not one-to-one over all real numbers, so to define inverses as functions we restrict domains.

• Arcsine: $y = \arcsin(x)$ means $\sin(y)=x$ with $y$ restricted to:

$-\frac{\pi}{2} \le y \le \frac{\pi}{2}$

• Arccosine: $y = \arccos(x)$ means $\cos(y)=x$ with $y$ restricted to:

$0 \le y \le \pi$

• Arctangent: $y = \arctan(x)$ means $\tan(y)=x$ with $y$ restricted to:

$-\frac{\pi}{2} < y < \frac{\pi}{2}$

These restrictions matter because derivatives depend on which branch you are using.

Why their derivatives look the way they do

A good way to understand inverse trig derivatives is to connect them to:

• the inverse function derivative rule, and
• implicit differentiation.

For instance, if $y=\arcsin(x)$ then $\sin(y)=x$. Differentiating implicitly relates $\frac{dy}{dx}$ to trig expressions in $y$, and then you convert everything back to $x$ using a triangle or identities.

This “differentiate in $y$, then translate to $x$” pattern is the core mechanism.

Derivative formulas you are expected to know

For differentiable inputs within the appropriate domains:

$\frac{d}{dx}\left(\arcsin(x)\right) = \frac{1}{\sqrt{1-x^2}}$

$\frac{d}{dx}\left(\arccos(x)\right) = -\frac{1}{\sqrt{1-x^2}}$

$\frac{d}{dx}\left(\arctan(x)\right) = \frac{1}{1+x^2}$

Also commonly used in AP Calculus BC:

\frac{d}{dx}\left(\arccot(x)\right) = -\frac{1}{1+x^2}

\frac{d}{dx}\left(\arcsec(x)\right) = \frac{1}{|x|\sqrt{x^2-1}}

\frac{d}{dx}\left(\arccsc(x)\right) = -\frac{1}{|x|\sqrt{x^2-1}}

The absolute value in the secant and cosecant inverses is important and often mishandled. It comes from converting $\tan(y)$-type ratios into square roots where sign can depend on the quadrant restrictions.

Deriving one formula (so it is not just memorization)

Derivative of $\arcsin(x)$

Let:

$y = \arcsin(x)$

That means:

$\sin(y) = x$

Differentiate implicitly with respect to $x$:

$\cos(y)\frac{dy}{dx} = 1$

So:

$\frac{dy}{dx} = \frac{1}{\cos(y)}$

Now you must rewrite $\cos(y)$ in terms of $x$. Since $\sin(y)=x$, you can use:

$\sin^2(y) + \cos^2(y) = 1$

So:

$x^2 + \cos^2(y) = 1$

$\cos^2(y) = 1 - x^2$

Because $y$ is restricted to $\left[-\frac{\pi}{2},\frac{\pi}{2}\right]$, $\cos(y)$ is nonnegative, so:

$\cos(y) = \sqrt{1-x^2}$

Therefore:

$\frac{d}{dx}\left(\arcsin(x)\right) = \frac{1}{\sqrt{1-x^2}}$

What could go wrong: If you forget the domain restriction, you might incorrectly use $\cos(y) = -\sqrt{1-x^2}$ in some cases. The restriction ensures the square root choice.

Using chain rule with inverse trig (most AP tasks)

In practice, you usually differentiate expressions like $\arctan(3x)$ or $\arcsin(x^2)$. The inverse trig derivative is the “outer derivative,” and then you multiply by the derivative of the inside.

Example 1: $\arcsin(x^2)$

Let:

$h(x) = \arcsin(x^2)$

Use the chain rule:

$h'(x) = \frac{1}{\sqrt{1-(x^2)^2}}\cdot 2x$

Simplify:

$h'(x) = \frac{2x}{\sqrt{1-x^4}}$

A frequent error here is to write $\sqrt{1-x^2}$ instead of $\sqrt{1-x^4}$ because you forgot to square the entire inside.

Example 2: Combining inverse trig with product/quotient rules

Differentiate:

$g(x) = x\arctan(x)$

Use product rule:

$g'(x) = 1\cdot \arctan(x) + x\cdot \frac{1}{1+x^2}$

So:

$g'(x) = \arctan(x) + \frac{x}{1+x^2}$

Inverse trig derivatives in related rates and slope problems

Because inverse trig functions return angles, they show up when you model an angle from a ratio (for example, angle of elevation). If an angle is defined by:

$\theta = \arctan\left(\frac{y}{x}\right)$

then differentiating $\theta$ requires the inverse trig derivative combined with quotient rule or implicit differentiation.

Example 3: A rate-of-change setup

Suppose:

$\theta(t) = \arctan\left(\frac{t}{2}\right)$

Then:

$\frac{d\theta}{dt} = \frac{1}{1+\left(\frac{t}{2}\right)^2}\cdot \frac{1}{2}$

Simplify:

$\frac{d\theta}{dt} = \frac{1}{2\left(1+\frac{t^2}{4}\right)}$

Multiply numerator and denominator by $4$:

$\frac{d\theta}{dt} = \frac{2}{t^2+4}$

This kind of simplification is common because AP free-response often expects a clean final form.

Exam Focus

• Typical question patterns:
  • Differentiate an expression containing $\arcsin$, $\arccos$, or $\arctan$ with a nontrivial inside function (chain rule).
  • Derive or justify a derivative using implicit differentiation (especially for $\arcsin$ and $\arctan$).
  • Use inverse trig derivatives inside related rates or motion contexts where an angle is defined by a ratio.
• Common mistakes:
  • Dropping the chain rule factor (forgetting to multiply by the derivative of the inside).
  • Incorrect radicand: writing $\sqrt{1-x^2}$ when the inside was something like $x^2$, leading to $\sqrt{1-x^4}$.
  • Ignoring the absolute value in \arcsec and \arccsc derivatives (or not recognizing when those forms are required).

Selecting Procedures for Calculating Derivatives

Why “procedure choice” is a skill (not just computation)

In Unit 3, you learn several derivative tools: chain rule for composites, implicit differentiation, and inverse-function ideas. On AP problems, the hard part is often not taking the derivative once you know what to do, but deciding which method fits the structure of the function you are given.

A helpful mindset is: before differentiating, ask “How is $y$ defined?”

• Is $y$ written explicitly as a formula in terms of $x$?
• Is $y$ defined implicitly by an equation involving both $x$ and $y$?
• Is $y$ defined as an inverse, like $f^{-1}(x)$ or $\arcsin(x)$?
• Is $y$ a composition of several layers?

The structure tells you which differentiation pathway will be shortest and least error-prone.

Procedure 1: Use the inverse derivative rule when you have values of $f$ and $f'$

If the problem gives you information about $f$ at some input (a table, a graph, a point and slope), and asks about $f^{-1}$, the inverse derivative rule is usually the intended approach.

Best fit clues:

• You see language like “Given $f(2)=5$ and $f'(2)=3$…”
• You are asked for $\left(f^{-1}\right)'(5)$.

Example 1: Choosing the inverse rule over algebra

Let $f(x) = x + \sin(x)$. Find $\left(f^{-1}\right)'(0)$.

Trying to solve $x+\sin(x)=0$ for $x$ exactly is not a good plan in general. Instead, use the point-matching idea.

First find $a$ such that $f(a)=0$. Notice:

$f(0) = 0 + \sin(0) = 0$

So $a=0$ and $b=0$.

Differentiate:

$f'(x) = 1 + \cos(x)$

Evaluate at $0$:

$f'(0) = 1 + \cos(0) = 2$

Apply inverse derivative rule:

$\left(f^{-1}\right)'(0) = \frac{1}{2}$

Why this was the right procedure: you could not reasonably write $f^{-1}(x)$ in elementary terms, but you did not need it.

Procedure 2: Use implicit differentiation when $y$ is not isolated

Sometimes the inverse rule is not directly visible, but the equation naturally defines an inverse relationship.

If you can rewrite an inverse situation as a composition identity or as an implicit equation, implicit differentiation can produce the derivative efficiently.

Best fit clues:

• The equation mixes $x$ and $y$ in a way that makes solving for $y$ messy.
• The equation resembles something like $f(y)=x$.

Example 2: Differentiating an inverse-defined relationship implicitly

Suppose $y = f^{-1}(x)$ and you also know:

$f(y) = y^3 - 4y$

Because $y=f^{-1}(x)$, you have:

$x = f(y) = y^3 - 4y$

Differentiate both sides with respect to $x$:

$1 = (3y^2 - 4)\frac{dy}{dx}$

So:

$\frac{dy}{dx} = \frac{1}{3y^2 - 4}$

If you want $\left(f^{-1}\right)'(x)$ explicitly in terms of $x$, you would need to replace $y$ with $f^{-1}(x)$:

$\left(f^{-1}\right)'(x) = \frac{1}{3\left(f^{-1}(x)\right)^2 - 4}$

This matches the general inverse derivative formula in a different form.

Common pitfall: forgetting that when you differentiate with respect to $x$, the derivative of $y$ is $\frac{dy}{dx}$, not $1$.

Procedure 3: Use inverse trig derivatives when the function is already an inverse trig

If the expression contains $\arcsin$, $\arccos$, $\arctan$, etc., you should usually apply the known inverse trig derivative formulas plus the chain rule.

Best fit clues:

• The “outer function” is an inverse trig function.
• You are not being asked to derive anything from scratch, just compute a derivative.

Example 3: Recognizing chain rule with inverse trig

Differentiate:

$p(x) = \arctan\left(x^2 + 1\right)$

Outer derivative:

$\frac{d}{dx}\left(\arctan(u)\right) = \frac{1}{1+u^2}$

Inner function:

$u = x^2 + 1$

$\frac{du}{dx} = 2x$

So:

$p'(x) = \frac{1}{1+\left(x^2+1\right)^2}\cdot 2x$

$p'(x) = \frac{2x}{1+\left(x^2+1\right)^2}$

Common pitfall: writing $\frac{1}{1+x^2}$ instead of $\frac{1}{1+u^2}$, which ignores the inside function.

Procedure 4: Use the chain rule perspective to avoid “formula hunting”

Many AP questions combine multiple ideas. A strong way to choose procedures is to identify layers:

• outer function
• inner function
• any implicit relationship
• any inverse relationship

If you see something like $f^{-1}(g(x))$, you can combine inverse derivative ideas with the chain rule.

Example 4: Chain rule with an inverse function

Assume $f$ is differentiable and invertible. Differentiate:

$q(x) = f^{-1}(x^2)$

Let $u=x^2$. Then:

$q'(x) = \left(f^{-1}\right)'(u)\cdot 2x$

Use the general inverse derivative formula:

$\left(f^{-1}\right)'(u) = \frac{1}{f'(f^{-1}(u))}$

Substitute back:

$q'(x) = \frac{2x}{f'(f^{-1}(x^2))}$

This is a common “compose with inverse” structure where you do not have numerical values, so you leave the result in terms of $f$ and $f^{-1}$.

How to decide quickly (without turning it into a checklist)

A good decision process is to translate the wording or structure into one of these templates:

• If you are asked for a derivative of $f^{-1}$ at a specific number, try to find the matching original point and use reciprocity.
• If the function is defined implicitly or you have $f(y)=x$, implicit differentiation is often the cleanest.
• If you see $\arcsin$ or friends, treat them as standard “outer functions” with known derivatives, then chain rule.
• If you are given a table/graph, resist inventing a formula; use the inverse derivative relationship with provided slope information.

Exam Focus

• Typical question patterns:
  • “Given a table of $f$ and $f'$, find $\left(f^{-1}\right)'(c)$” (procedure selection is the main point).
  • Differentiate a function that combines inverse trig with algebraic layers (chain rule plus inverse trig derivative).
  • Differentiate an expression like $f^{-1}(g(x))$ and leave the answer in terms of $f'$ and $f^{-1}$.
• Common mistakes:
  • Choosing algebraic inversion when it is unnecessary (wastes time and increases error risk).
  • Using the reciprocal rule with the wrong input (not matching $f(a)=b$ correctly).
  • Mixing methods incorrectly, such as trying to apply inverse trig derivative formulas to a regular trig function without recognizing it is not an inverse.