Why Definite-Integral Errors Happen
A definite integral produces a number that represents accumulated change over an interval. Even when an antiderivative is correct, a small error in applying the bounds can change that number: reversing the subtraction changes its sign, using endpoints for the wrong variable gives the wrong values, and omitting the lower endpoint leaves out part of the calculation.
The safest habit is to keep the evaluation visible until the final arithmetic is complete. The notation \([F(x)]_a^b\) is not a reminder to substitute one endpoint. It means evaluate at the upper endpoint and subtract the value at the lower endpoint.
What the Bracket Notation Means
For an antiderivative \(F\), the notation \([F(x)]_a^b\) is defined by \([F(x)]_a^b=F(b)-F(a)\). The upper limit is written on top, but the order of subtraction is fixed: upper value minus lower value.
This order matters even when the lower endpoint is negative, or when \(F(a)\) itself is negative. Subtract the entire lower-endpoint value, including its sign. Parentheses make the intended operation clear.
The evaluation rule requires an antiderivative on the interval and a definite integral for which the Fundamental Theorem of Calculus applies. In the usual AP examples, the integrand is continuous on the closed interval and the antiderivative is differentiable there. Always check that the function being integrated and the interval make the integral meaningful.
Why the Endpoint Difference Works
The rule follows from the Fundamental Theorem of Calculus. Let \(f\) be continuous on \([a,b]\), and suppose \(F'=f\) there. Define the accumulated integral \(A(x)=\int_a^x f(t)\,dt\). The theorem gives \(A'(x)=f(x)\). Since \(F'(x)=f(x)\) as well, the difference \(F(x)-A(x)\) has derivative zero on the interval.
By the Mean Value Theorem, a function with derivative zero on an interval is constant there. At \(x=a\), \(A(a)=\int_a^a f(t)\,dt=0\), so \(F(a)-A(a)=F(a)\). Therefore \(F(x)-A(x)=F(a)\), or \(A(x)=F(x)-F(a)\). Setting \(x=b\) gives the evaluation rule.
In short, the lower-endpoint value is required because the accumulated integral starts at zero when \(x=a\). Subtracting \(F(a)\) shifts the antiderivative’s values so that the change from \(a\) to \(b\), rather than its absolute value at \(b\), is measured.
Worked Example: Reversed Bounds
Worked Example: Reversed Bounds
Evaluate \(\int_3^1 (2x+1)\,dx\). The lower limit is 3 and the upper limit is 1, so the bounds are reversed relative to the usual increasing order.
An antiderivative is \(F(x)=x^2+x\), because \(\frac{d}{dx}(x^2+x)=2x+1\). Apply the bounds in their written order:
Check the sign by reversing the bounds. The integral in the opposite direction is \(\int_1^3(2x+1)\,dx=12-2=10\), so \(\int_3^1(2x+1)\,dx=-10\), as required by \(\int_b^a f(x)\,dx=-\int_a^b f(x)\,dx\). The antiderivative check also holds: \(F'(x)=2x+1\).
A frequent error is to see \(3\) as the “upper” endpoint because it is larger. The endpoint roles are set by the integral’s notation, not by which number is greater. In \(\int_3^1\), the upper limit is 1 and the lower limit is 3.
Worked Example: Substitution and Endpoint Variables
Worked Example: Substitution and Endpoint Variables
Evaluate \(\int_0^2 2x e^{x^2}\,dx\). Let \(u=x^2\), so \(du=2x\,dx\). Because the integration variable changes to \(u\), the endpoints must also be expressed as \(u\)-values.
When \(x=0\), \(u=0^2=0\). When \(x=2\), \(u=2^2=4\). Thus the transformed integral is \(\int_0^4 e^u\,du\), not \(\int_0^2 e^u\,du\). An antiderivative of \(e^u\) is \(e^u\), and evaluation gives
To check the result using \(x\), an antiderivative of the original integrand is \(F(x)=e^{x^2}\), since \(\frac{d}{dx}e^{x^2}=e^{x^2}(2x)=2x e^{x^2}\). Then \(F(2)-F(0)=e^4-1\), agreeing with the transformed calculation.
A common wrong calculation uses the new integrand \(e^u\) but keeps the old endpoints \(0\) and \(2\), producing \(e^2-1\). That calculation mixes variables: \(0\) and \(2\) are \(x\)-values, while the transformed integral requires \(u\)-values. Alternatively, one may keep the original \(x\)-bounds by returning to \(x\) and using \(F(x)=e^{x^2}\).
There are two consistent approaches to substitution in a definite integral: change the variable and transform both bounds, or find an antiderivative in the new variable, return to the original variable, and then use the original bounds. The error is not choosing one approach over the other; it is mixing their variables and endpoints.
Worked Example: Subtracting the Lower Endpoint
Worked Example: Subtracting the Lower Endpoint
Evaluate \(\int_1^2 (x^2+1)\,dx\). An antiderivative is \(F(x)=\frac{x^3}{3}+x\), because
Now evaluate at both limits and subtract the lower-endpoint value:
The derivative check confirms the antiderivative, and the endpoint difference is \(\frac{10}{3}\). Evaluating only at the upper endpoint would give \(\frac{14}{3}\), which is not the definite integral. The missing \(\frac{4}{3}\) is \(F(1)\), the value that must be subtracted.
Common Wrong Calculations and Their Corrections
These errors can be caught by writing the endpoint evaluations explicitly before simplifying. In particular, keep parentheses around the entire lower-endpoint value:
If \(F(a)\) is negative, subtracting it adds its magnitude. For example, if \(F(b)=5\) and \(F(a)=-2\), then \(F(b)-F(a)=5-(-2)=7\), not \(5-2=3\).
| Wrong move | Correction |
|---|---|
| Subtracting the upper value from the lower value | Use upper value minus lower value: \(F(b)-F(a)\). |
| Changing the limits’ order but keeping the integral’s sign | Reversing the limits reverses the sign. |
| Using original-variable endpoints after changing the integration variable | Transform both endpoints, or return to the original variable before evaluating. |
| Evaluating only at the upper endpoint | Subtract the lower-endpoint value, even if it is not zero. |
| Dropping parentheses around a negative lower value | Write \(F(b)-\bigl(F(a)\bigr)\) before simplifying. |
A Reliable Evaluation Routine
Use the following routine whenever you evaluate a definite integral with an antiderivative. It makes the variable, the endpoint order, and the subtraction visible.
Choose \(F\) with \(F'=f\), and differentiate \(F\) to check that it returns the integrand.
If a substitution changes the variable, either convert both bounds or return the antiderivative to the original variable.
Use \([F(x)]_a^b=F(b)-F(a)\), preserving the written upper and lower limits.
Use parentheses, especially when an endpoint value is negative, and then simplify the difference.
Confirm the antiderivative by differentiation and consider whether the sign and endpoint order are consistent.
Why Careful Evaluation Matters
Definite integrals are used to measure net change, signed area, and accumulated quantities. A sign error can turn a decrease into an increase, and a missing lower-endpoint value can make a correct antiderivative produce an incorrect total. These are not errors in integration technique; they occur after the antiderivative has already been found.
When a result seems suspicious, return to the written bounds and the bracket notation rather than repeating the antiderivative calculation. Ask: Which endpoint is upper in the integral as written? Did I evaluate \(F\) at both endpoints? Did I subtract the lower value, including its sign? If substitution was used, are the endpoints expressed in the same variable as the transformed integral?
Check Your Understanding
For each question, show the endpoint substitution before simplifying.
- Evaluate \(\int_4^1 2x\,dx\). Explain the sign of your result.
- For \(\int_0^3 2x\cos(x^2)\,dx\), use \(u=x^2\). What are the transformed bounds?
- Evaluate \(\int_1^2 (3x^2+1)\,dx\). What incorrect value results if you use only the upper endpoint?
- If \(F(5)=2\) and \(F(1)=-4\), what is \([F(x)]_1^5\)?
- State the relationship between \(\int_b^a f(x)\,dx\) and \(\int_a^b f(x)\,dx\).
