What Are Monotonic Functions?
Monotonic Functions are functions that keep a single direction of change on an interval. On that interval the function is either non-decreasing (its outputs never go down as the input rises) or non-increasing (its outputs never go up as the input rises). A function that rises for a while and then falls is not monotonic on the whole interval, even though it is monotonic on each piece separately.
Fix an interval $I$ and take any two inputs $x_1 < x_2$ in it. The four precise definitions are:
Non-decreasing (increasing): $f(x_1) \le f(x_2)$. The output never drops.
Strictly increasing: $f(x_1) < f(x_2)$. The output always rises, with no flat steps.
Non-increasing (decreasing): $f(x_1) \ge f(x_2)$. The output never rises.
Strictly decreasing: $f(x_1) > f(x_2)$. The output always falls.
A function is monotonic on $I$ when one of these holds for every pair of inputs in $I$. The difference between the weak inequality ($\le$) and the strict one ($<$) is the single most tested idea in the topic, and the increasing and decreasing intervals of a function are exactly the intervals on which one of these four conditions is true.
The geometric reading is simpler than the symbols. A derivative is the slope of the tangent line at a point. If every tangent along the curve tilts upward, the curve climbs, and the function is increasing.
If every tangent tilts downward instead, the curve falls, and the function is decreasing. Monotonicity is just the statement that the tangent lines all lean the same way.
How Do You Test A Function For Monotonicity?
To test monotonicity you look at the sign of the first derivative across the interval. For a function differentiable on an open interval:
$$f'(x) > 0 \ \text{on } I \ \Rightarrow \ f \text{ is strictly increasing on } I$$ $$f'(x) < 0 \ \text{on } I \ \Rightarrow \ f \text{ is strictly decreasing on } I$$ $$f'(x) = 0 \ \text{at a point} \ \Rightarrow \ \text{stationary candidate (test further)}$$
The precise two-way statement uses the weak inequality: $f$ is non-decreasing on $I$ exactly when $f'(x) \ge 0$ for all $x$ in $I$, and non-increasing exactly when $f'(x) \le 0$ throughout. Strict positivity of the derivative guarantees strict increase, but it is not the only way to get it, a point the first worked example makes concrete.
The working method is a sign chart:
Find $f'(x)$.
Solve $f'(x) = 0$ and note where $f'(x)$ is undefined. These split the domain into test intervals.
Pick one sample point inside each interval and record the sign of $f'$ there.
Read off increasing intervals (where $f' > 0$) and decreasing intervals (where $f' < 0$).
The points from step 2 are the stationary and critical points. They are the only places a monotonic run can turn around.
How Do You Find Intervals Of Increase And Decrease? (Worked Example)
Example 1: A stationary point that does not break monotonicity, $f(x) = x^3$.
Differentiate:
$$f'(x) = 3x^2$$
Now solve $f'(x) = 0$: the only solution is $x = 0$. Everywhere else $3x^2 > 0$. So the derivative is zero at a single isolated point and positive on both sides of it.
Does that stationary point split $x^3$ into an increasing piece and a decreasing piece? No. Check directly with two inputs $x_1 < x_2$: cubing preserves order for all real numbers, so $x_1^3 < x_2^3$ always holds. The function is strictly increasing on all of $\mathbb{R}$, even though $f'(0) = 0$.
Final answer: $x^3$ is strictly increasing on $(-\infty, \infty)$; the stationary point at $x = 0$ is a flat tangent, not a turning point.
The lesson: $f' > 0$ everywhere is sufficient for strict increase, but not necessary. A derivative may touch zero at isolated points and the function still climbs the whole way.
Example 2: A genuine turning point, $f(x) = x^2$.
Differentiate:
$$f'(x) = 2x$$
Solve $f'(x) = 0$: this gives $x = 0$. Test each side:
For $x < 0$, say $x = -1$: $f'(-1) = -2 < 0$, so $f$ is decreasing on $(-\infty, 0)$.
For $x > 0$, say $x = 1$: $f'(1) = 2 > 0$, so $f$ is increasing on $(0, \infty)$.
The sign genuinely changes at $x = 0$, so this stationary point is a real turning point (a minimum).
Final answer: $x^2$ is decreasing on $(-\infty, 0)$ and increasing on $(0, \infty)$. It is monotonic on each of those intervals but not monotonic on the whole real line.
Example 3: A rational function via a full sign chart, $f(x) = \dfrac{x}{x^2 + 1}$.
Differentiate with the quotient rule:
$$f'(x) = \frac{(x^2+1)(1) - x(2x)}{(x^2+1)^2} = \frac{x^2 + 1 - 2x^2}{(x^2+1)^2} = \frac{1 - x^2}{(x^2+1)^2}$$
The denominator $(x^2+1)^2$ is positive for every real $x$, so the sign of $f'$ is the sign of the numerator $1 - x^2$. Solve $1 - x^2 = 0$ to get $x = -1$ and $x = 1$. Build the sign chart:
Table: Sign chart of $f'(x) = \dfrac{1 - x^2}{(x^2+1)^2}$.
Interval | Sample $x$ | Sign of $1 - x^2$ | $f'(x)$ | Behaviour |
|---|---|---|---|---|
$(-\infty, -1)$ | $x = -2$ | $1 - 4 = -3$ | negative | decreasing |
$(-1,\ 1)$ | $x = 0$ | $1 - 0 = 1$ | positive | increasing |
$(1, \infty)$ | $x = 2$ | $1 - 4 = -3$ | negative | decreasing |
Final answer: $f(x) = \dfrac{x}{x^2+1}$ is decreasing on $(-\infty, -1)$, increasing on $(-1, 1)$, and decreasing on $(1, \infty)$. The stationary point at $x = -1$ is a local minimum with value $f(-1) = -\tfrac{1}{2}$, and the one at $x = 1$ is a local maximum with value $f(1) = \tfrac{1}{2}$.
How Do Monotonic Functions Connect To Inverses?
A strictly monotonic function is always one-to-one, and being one-to-one is precisely what lets a function be inverted. The chain runs in three links:
Strictly monotonic implies one-to-one. If $f$ is strictly increasing and $x_1 \ne x_2$, then whichever is smaller has the smaller output, so $f(x_1) \ne f(x_2)$. Different inputs never collide on the same output, which is the definition of a one-to-one function.
One-to-one implies invertible. A one-to-one function pairs each output with exactly one input, so the pairing can be run backwards. That backward pairing is the inverse function.
The inverse keeps the direction. The inverse of a strictly increasing function is itself strictly increasing.
This is why $f(x) = x^3$ has the clean inverse $f^{-1}(x) = x^{1/3}$: it was strictly increasing, so nothing was lost. The natural logarithm gives the same story. Because $f(x) = \ln x$ has $f'(x) = \frac{1}{x} > 0$ for every $x > 0$, it is strictly increasing on its domain, hence one-to-one, hence invertible, and its inverse is $e^x$. More on that family lives in logarithmic functions.
The contrast is $f(x) = x^2$ on the whole real line. It is not monotonic there, and it fails the one-to-one test, since $f(-2) = f(2) = 4$. Only after you restrict it to $[0, \infty)$, where it is strictly increasing, does the inverse $\sqrt{x}$ appear. Monotonicity is the gate to invertibility.
Why Does The First Derivative Test Work?
The sign of the derivative controls the direction of the function because of the Mean Value Theorem, one of the central results in calculus. The intuition is worth stating plainly.
The derivative $f'(c)$ is the slope of the tangent at a single point $c$, the instantaneous rate of change.
Over an interval from $x_1$ to $x_2$, the average rate of change is the slope of the straight line (the secant) joining the two endpoints.
The Mean Value Theorem says that if $f$ is continuous on $[x_1, x_2]$ and differentiable on $(x_1, x_2)$, then some interior point $c$ has a tangent slope equal to that secant slope: $f'(c) = \dfrac{f(x_2) - f(x_1)}{x_2 - x_1}$.
Now suppose $f'(x) > 0$ across the interval. Then that interior slope $f'(c)$ is positive, so the right-hand side is positive. Since the denominator $x_2 - x_1$ is positive, the numerator $f(x_2) - f(x_1)$ must be positive too, which says $f(x_2) > f(x_1)$.
Every pair of inputs obeys this, so $f$ is strictly increasing. Flip the inequality and the same argument delivers strictly decreasing.
That is the whole reason the test is trustworthy. It is not a rule of thumb. If every tangent slopes upward, the theorem forces every secant to slope upward too, and a function whose every secant rises cannot turn back down.
Who Discovered The Link Between Slope And Monotonic Functions?
The idea that the sign of a rate of change decides whether a quantity grows is as old as calculus, but turning it into a theorem took the generation that made calculus rigorous.
Two named figures anchor the modern story: Joseph-Louis Lagrange (1736–1813, born in Turin, Italy) for the Mean Value Theorem, and Augustin-Louis Cauchy (1789–1857, France) for the rigorous limit foundations that made the monotonicity test airtight.
Where Are Monotonic Functions Used In The Real World?
Monotonicity is the property that lets a quantity be reversed, searched, or trusted to move one way, so it turns up across science and engineering.
Computer science: binary search only works on data that is sorted, which is a monotonic arrangement. The whole speed of the method rests on the list running one way.
Machine learning: models used for credit scoring or pricing are often given monotonic constraints, so that a higher income can never lower a predicted score. Isotonic regression fits a monotonic curve to noisy data on purpose.
Statistics: a cumulative distribution function is non-decreasing by construction, rising from $0$ to $1$, and rank correlation measures how close a relationship is to being monotonic.
Engineering and instrumentation: a sensor calibration curve must be monotonic to be invertible, otherwise a single reading could map back to two different true values and the instrument would be ambiguous.
Economics and motion: an object that only ever moves forward has a non-decreasing position, and a demand curve is usually modelled as monotonic in price.
Across all of these, the shared reason is the same one from the inverse section: a monotonic relationship can be run backwards without confusion, and much of applied mathematics depends on that.
What Are The Most Common Mistakes With Monotonic Functions?
These four errors account for most lost marks on monotonicity, verified against the definition warnings on Wikipedia's monotonic-function article and the reader questions surfaced in Vedantu's and the JEE prep FAQ lists.
Confusing "strictly increasing" with "non-decreasing."
Where it slips in:
A student treats $f(x_1) \le f(x_2)$ and $f(x_1) < f(x_2)$ as the same condition and calls a function that stays flat over a stretch "strictly increasing."
Don't do this:
Do not ignore the flat steps. A function that is constant on part of an interval is non-decreasing there, but it is not strictly increasing, because equal outputs are allowed under $\le$ and forbidden under $<$.
The correct way:
Check whether equality can occur. If the outputs can repeat, the function is non-decreasing but not strict. A constant function is both non-decreasing and non-increasing, yet neither strictly increasing nor strictly decreasing.
Reading "non-decreasing" as "not decreasing."
Where it slips in:
A student meets a function that first falls and then rises and labels it non-decreasing because "it is not always decreasing."
Don't do this:
Do not confuse "non-decreasing" with the far weaker "not decreasing." A function that dips and then climbs is not decreasing and not increasing, but it is neither non-decreasing nor non-increasing.
The correct way:
Use the definition on every pair of inputs. Non-decreasing means no pair ever drops. One place where a later input has a smaller output breaks it.
Treating a zero derivative as an automatic turning point.
Where it slips in:
A student finds $f'(a) = 0$ and immediately declares that the function changes direction at $a$, so $f(x) = x^3$ gets wrongly split at $x = 0$.
Don't do this:
Do not stop at $f'(a) = 0$. A stationary point is only a candidate for a turn, not a guarantee of one.
The correct way:
Check the sign of $f'$ on both sides. If it stays positive across the point, as with $3x^2$ at $x = 0$, the function keeps increasing and the point is a flat tangent, not a turning point.
Claiming monotonicity across a broken domain.
Where it slips in:
A student computes $f'(x) < 0$ for $f(x) = \dfrac{1}{x}$ and concludes the function is decreasing on its whole domain.
Don't do this:
Do not stitch two separate intervals into one monotonic claim. Although $f'(x) = -\dfrac{1}{x^2} < 0$ wherever it is defined, the domain is split at $x = 0$.
The correct way:
State monotonicity per interval. $\dfrac{1}{x}$ is strictly decreasing on $(-\infty, 0)$ and strictly decreasing on $(0, \infty)$, but not monotonic across the union, since $f(-1) = -1$ is less than $f(1) = 1$.
Practice Problems On Monotonic Functions
Verify each by finding $f'(x)$ and reading its sign. Answers follow each problem.
Is $f(x) = 5x - 2$ monotonic on $\mathbb{R}$?
(Answer: $f'(x) = 5 > 0$, so it is strictly increasing everywhere, hence monotonic and invertible.)Find the intervals of increase and decrease for $f(x) = x^2 - 4x + 1$.
(Answer: $f'(x) = 2x - 4 = 0$ at $x = 2$; decreasing on $(-\infty, 2)$, increasing on $(2, \infty)$.)Show that $f(x) = x^3 + x$ is strictly increasing on $\mathbb{R}$.
(Answer: $f'(x) = 3x^2 + 1 \ge 1 > 0$ for all $x$, so it is strictly increasing, therefore one-to-one and invertible.)Find the intervals of increase and decrease for $f(x) = x^3 - 3x$.
(Answer: $f'(x) = 3x^2 - 3 = 3(x-1)(x+1)$; increasing on $(-\infty, -1)$ and $(1, \infty)$, decreasing on $(-1, 1)$.)Is $f(x) = e^{-x}$ increasing or decreasing?
(Answer: $f'(x) = -e^{-x} < 0$ for all $x$, so it is strictly decreasing on $\mathbb{R}$, hence invertible.)Determine the monotonic behaviour of $f(x) = \dfrac{1}{x}$.
(Answer: $f'(x) = -\dfrac{1}{x^2} < 0$; strictly decreasing on $(-\infty, 0)$ and on $(0, \infty)$ separately, but not monotonic on the whole domain.)
Where Should You Go Next After Monotonic Functions?
Monotonicity is a hinge topic in the application of derivatives, and several natural doors open from here.
Increasing and decreasing intervals. Drill the sign-chart method on more functions, including ones where the derivative is undefined rather than zero.
Inverse functions. Follow the one-to-one chain forward and learn how to build and check an inverse once monotonicity has guaranteed one exists.
The derivative. Firm up the tool the whole test rests on, from the slope-of-tangent idea to the rules for differentiating.
If your child is building these foundations, a live Bhanzu trainer teaches monotonicity starting from the "why" (the Mean Value Theorem link between slope and direction) with a Bhanzu math tutor.
Was this article helpful?
Your feedback helps us write better content