Integrals require antiderivatives. Antiderivatives require undoing derivative rules.
This section shows you how to undo the chain rule. The method is called U-Substitution!
Recall from Section 2.9 the definition of a differential:
After picking your $u$, you need to replace the $dx$ in your integral! The differential $du$ allows you to do this replacement.
First, we will rewrite the integral: \[\int x^3 \cos(x^4 + 2) \ dx = \int \cos(x^4 + 2) \cdot x^3 \ dx\]
Let $u = x^4 + 2$, the inside of the composition.
Then $du = 4x^3 \ dx$.
Look at your integral. You need to replace $x^3 \ dx$. Solving for this in the previous equation: $x^3 \ dx = \dfrac{1}{4} \ du$.
Now substitute:
\begin{align} \int \cos(x^4 + 2) \cdot x^3 \ dx &= \int cos(u) \dfrac{1}{4} \ du \\&= \dfrac{1}{4} \int \cos(u) \ du \\&= \dfrac{1}{4} \left(\sin(u) + C\right) \\&= \dfrac{\sin{u}}{4} + \dfrac{1}{4}C \\&= \dfrac{\sin(x^4 + 2)}{4} + C \end{align}Notice that I replaced $\dfrac{1}{4}C$ with $C$. You can do this because it's just a constant, it doesn't matter if $C$ is divided by four or not.
Recall in Section 4.2 we said the $dx$ is symbolically equivalent to the width of one rectangle.
In this section we are manipulating $dx$ as part of a differential.
We are allowed to do this because of the following reasons:
Using U-substition on definite integrals is the same method as indefinite integrals.
The only difference is the top and bottom bound of integration must be changed from $x$ to $u$'s.