From Endpoint Outputs to Every Value Between
A continuous function can rise, fall, or turn around. But on an interval, it cannot move from one output level to another while skipping a level in between. This is the central interpretation of the Intermediate Value Theorem: if the function is continuous on a closed interval, then every value between its endpoint outputs occurs somewhere on that interval.
To use the theorem, focus on the function’s outputs at the endpoints—not merely on the endpoint inputs. If \(f(a)=2\) and \(f(b)=7\), for example, the theorem guarantees that each value from \(2\) to \(7\) is attained, provided \(f\) is continuous on \([a,b]\).
The Intermediate Value Theorem
Continuity on a closed interval means that \(f\) is continuous at every point of \([a,b]\), including its endpoints in the one-sided sense appropriate to the domain. The Intermediate Value Theorem then connects the endpoint outputs to all the outputs between them.
“Between” allows either order of the endpoint outputs. If \(f(a)<f(b)\), the possible target values satisfy \(f(a)\le y\le f(b)\). If \(f(b)<f(a)\), they satisfy \(f(b)\le y\le f(a)\). The function does not need to be increasing or decreasing, and the theorem does not say that the input \(c\) is unique.
The conclusion follows directly from the theorem’s hypotheses: check continuity on the whole closed interval, calculate the two endpoint outputs, and verify that the desired \(y\) lies between them. The theorem then supplies an input \(c\) in the interval with output \(y\). If \(y\) equals an endpoint output, an endpoint itself may provide that input; a target strictly between the outputs must be reached at an interior point.
What the Graph Tells You
Imagine the graph of \(y=f(x)\) drawn from \(x=a\) to \(x=b\). The points at the ends have heights \(f(a)\) and \(f(b)\). Choose a horizontal line \(y=y_0\) whose height lies between those two heights. The theorem says that the graph meets this horizontal line at some point with input \(c\in[a,b]\). At that point, \(f(c)=y_0\).
This picture is a useful interpretation, not a replacement for checking the conditions. A gap, jump, or break can allow a graph to miss a level between its endpoint heights. The guarantee applies when the function is continuous on the entire interval, not just at its endpoints or at most points.
The theorem also does not require the graph to travel directly from one endpoint to the other. It may oscillate or turn back. That can make a particular output occur more than once, but the theorem only guarantees that it occurs at least once.
Worked Example: A Straight-Line Graph
Let \(f(x)=2x+1\) on \([0,3]\). Does \(f\) attain the output \(4\) on this interval?
First, \(f\) is a linear function, so it is continuous on \([0,3]\). Calculate the endpoint outputs:
The target \(4\) lies between \(1\) and \(7\), since \(1<4<7\). The Intermediate Value Theorem guarantees some \(c\in(0,3)\) with \(f(c)=4\). In this example we can find it by solving the equation:
Check that this input is in the interval and gives the target output:
The graph meets the horizontal line \(y=4\) at input \(x=\frac{3}{2}\), exactly as the theorem guarantees.
Finding a Guaranteed Output Does Not Require Solving
The theorem is useful even when the input \(c\) cannot be found by simple algebra. Its conclusion is an existence statement: it guarantees that at least one suitable input lies in the interval. To apply it, you do not need a formula for that input.
When an exact input is easy to calculate, substituting it back checks the result. When it is not easy to calculate, the guarantee still follows once continuity and the endpoint-output comparison have been verified. The theorem gives existence, not necessarily a method for locating the input exactly.
Worked Example: A Value Between the Outputs of a Square Function
Let \(f(x)=x^2\) on \([0,2]\). Show that the function attains the output \(2\).
The function \(x^2\) is continuous on \([0,2]\). Its endpoint outputs are
Since \(0<2<4\), the target output is strictly between the endpoint outputs. The theorem guarantees a \(c\in(0,2)\) such that \(c^2=2\). Solving this equation gives \(c=\sqrt{2}\) or \(c=-\sqrt{2}\). Only the positive value belongs to \([0,2]\): indeed, \(1<\sqrt{2}<2\), whereas \(-\sqrt{2}<0\).
Check the selected input and substitute it into the function:
Thus \(x=\sqrt{2}\) is in the interval and gives the required output. The theorem guarantees that an input exists; solving the equation identifies one in this particular example.
A Continuous Function Can Turn Around
Do not picture the theorem as a claim that the function always moves in one direction. A graph can turn around and still attain every level between its endpoint heights. Continuity, rather than monotonicity, is the key condition.
Worked Example: A Function That Turns Around
Consider \(f(x)=x^2-2x\) on \([0,3]\). Show that the function attains the output \(1\), even though its graph turns around.
This polynomial is continuous on \([0,3]\). Calculate its endpoint outputs:
The target \(1\) is between the endpoint outputs, because \(0<1<3\). The theorem therefore guarantees at least one \(c\in(0,3)\) with \(f(c)=1\). To identify such an input in this example, solve:
Using the quadratic formula gives
Since \(1<\sqrt{2}<2\), the value \(1+\sqrt{2}\) is greater than \(0\) and less than \(3\). The other value, \(1-\sqrt{2}\), is less than \(0\), so it lies outside the interval. Check the input in the interval by substitution:
Thus \(c=1+\sqrt{2}\) is a valid input in \([0,3]\) giving output \(1\). The turning shape of the graph does not prevent it from attaining the target.
A Common Mistake: Comparing the Wrong Numbers
The theorem compares the target output with \(f(a)\) and \(f(b)\). It does not compare the target with the inputs \(a\) and \(b\). Confusing those pairs can produce an invalid conclusion.
Calculate those outputs instead:
The target \(0\) is not between \(2\) and \(5\). Therefore, the Intermediate Value Theorem does not guarantee an input with output \(0\). In fact, for every real \(x\), \(x^2\ge 0\), so \(x^2+1\ge 1\); there is no such input at all.
A reliable check is to write down the endpoint outputs before comparing them with the target. Also verify that the function is continuous on the entire closed interval. Without both checks, the theorem cannot be applied.
What the Theorem Does and Does Not Say
- It guarantees an attained output: every value between the endpoint outputs occurs somewhere on the interval.
- It requires continuity on the whole closed interval: continuity only at the endpoints is not enough.
- It allows either endpoint-output order: the output at \(b\) may be larger or smaller than the output at \(a\).
- It guarantees existence, not uniqueness: there may be more than one input with the same output.
- It does not assert that every real number is attained: the guarantee concerns only values between \(f(a)\) and \(f(b)\).
In applications, this is a way to establish that a continuous model must pass through a specified level. In a graph, it guarantees an intersection with a horizontal line whose height is between the endpoint heights. The key work is checking the hypotheses and comparing the correct output values.
Check Your Understanding
For each question, identify the interval, the endpoint outputs, and whether the Intermediate Value Theorem applies.
- If \(f\) is continuous on \([a,b]\), with \(f(a)=6\) and \(f(b)=2\), which outputs does the theorem guarantee are attained?
- Why must the target be compared with \(f(a)\) and \(f(b)\), rather than with \(a\) and \(b\)?
- For \(f(x)=x^2\) on \([0,3]\), does the theorem guarantee that output \(5\) occurs? State the endpoint outputs and explain.
- Does the theorem guarantee a unique input for each output between \(f(a)\) and \(f(b)\)?
- What continuity condition must be checked before applying the theorem on \([a,b]\)?
