Increasing and Decreasing Intervals
A function is increasing on an interval when larger inputs give larger outputs, its graph rises left to right, and decreasing when the graph falls. Calculus reduces the question to a sign check:
f′ > 0 on an interval ⇒ f increasing there; f′ < 0 ⇒ f decreasing
Since a continuous derivative can only change sign by passing through zero, the real line splits at the critical points into stretches of constant behavior, and one test value per stretch settles everything. That split-and-test routine is the sign chart, the workhorse of curve analysis, and this calculator automates it: it differentiates symbolically, finds the critical points in your window, tests each resulting interval, and reports the rising and falling stretches with the evidence, drawing f and f′ together so the correspondence is visible.
How to Use the Increasing/Decreasing Intervals Calculator
Enter a function and a window to analyze. The steps show f′, the critical points, and one line per interval: the test point used, the sign of f′ there, and the verdict. The result line compresses everything into arrows, ↑ and ↓ with interval notation. On the graph, the solid curve is f and the dashed curve is f′: wherever the dashed curve sits above the axis, the solid one climbs, and the crossing points of the dashed curve are exactly the peaks and valleys of the solid one. That picture, once internalized, is most of curve sketching; the curve sketching calculator adds concavity and asymptotes to complete it.
Worked Example
Analyze f(x) = x³ − 3x. The derivative f′(x) = 3x² − 3 = 3(x − 1)(x + 1) vanishes at ±1, splitting the line into three intervals:
on (−∞, −1): f′(−2) = 9 > 0, increasing; on (−1, 1): f′(0) = −3 < 0, decreasing; on (1, ∞): f′(2) = 9 > 0, increasing
The pattern up-down-up immediately locates the local max at x = −1 and the local min at x = 1, no second derivative needed. The x + sin(x) chip shows a subtler case: f′ = 1 + cos(x) touches zero at odd multiples of π but never goes negative, so the function is increasing on the whole line despite momentary flat spots, a useful reminder that isolated zeros of f′ do not break monotonicity.
Why One Test Point Per Interval Suffices
The justification is the Intermediate Value Theorem applied to f′: a derivative that is, say, positive at one point of an interval and negative at another would have to vanish in between (derivatives have the intermediate value property by Darboux's theorem, even when discontinuous), contradicting that all zeros were collected as endpoints of the split. So the sign at one sample is the sign throughout. The logic silently assumes every sign-change location was found, which is why the calculator hunts zeros densely, and why by hand you must also include points where f′ is undefined: for f = x^(2/3), the derivative blows up at 0 and the monotonicity flips there without f′ ever equaling zero. Critical points of both kinds, zeros and non-existence points, are the complete list of possible switches, the same list that feeds the first derivative test.
Common Mistakes to Avoid
- Testing the critical points instead of the intervals. f′ is zero at the split points by construction; the information lives between them. Pick interior test values.
- Merging intervals across a critical point. "Increasing on (−∞, −1) ∪ (1, ∞)" invites error: the union statement is false for the function as a whole in general use. Report each interval separately.
- Forgetting undefined points of f or f′. 1/x decreases on each side of 0 but is not "decreasing on its domain" as one interval; the break at 0 must appear in the chart even though f′ = −1/x² is never zero.
- Reading isolated zeros of f′ as direction changes. x³ has f′(0) = 0 yet increases through the origin; a switch requires a genuine sign change, not merely a zero.
- Confusing f increasing with f positive. The sign of f′ controls direction; the sign of f controls position. A function can be negative and increasing (climbing toward zero) or positive and decreasing.
Real-World Applications
Monotonicity analysis answers "when is it getting better, and when worse." Economics locates the production level where marginal profit turns negative, the boundary between the increasing and decreasing stretches of the profit curve, which is exactly where optimization stops. Medicine reads drug-concentration curves: the increasing phase is absorption, the decreasing phase elimination, and the switch point is peak concentration, all first-derivative sign facts. Epidemiology's "the wave is slowing" is a statement that a case curve's derivative changed sign; climate science tracks intervals where temperature trends rise. Engineering uses monotonicity to verify safety margins, a stress response that increases monotonically with load has no hidden reversals, and mathematics itself uses it constantly: proving a function increasing is the standard route to proving inequalities and invertibility, since a strictly monotone function is automatically one-to-one, the entry ticket to the inverse function.
Frequently Asked Questions
How do I find where a function is increasing?
Differentiate, find every critical point (f′ zero or undefined), split the domain there, and test f′'s sign once per piece: positive means increasing on that whole piece. The calculator's step list is exactly this procedure with the arithmetic shown.
Why do the intervals split exactly at critical points?
A derivative can only change sign where it is zero or undefined, by Darboux's intermediate-value property of derivatives. Collect those points and no sign change can hide inside any resulting piece, which is what licenses the one-test-point shortcut.
Should intervals be open or closed at the endpoints?
Conventions differ and both are defensible: monotonicity is often stated on open intervals, yet including an endpoint where f is continuous is also correct (x² is decreasing on (−∞, 0] under the definition). What matters is consistency; this tool reports open intervals between critical points.
Can a function be increasing at a single point?
The useful notion is interval-based: f′(c) > 0 guarantees increase on some neighborhood of c only when f′ is continuous. Pathological examples exist with positive derivative at a point yet no monotone neighborhood; that is why analyses run on intervals, not points.
What if f′ is zero on a whole subinterval?
Then f is constant there, neither increasing nor decreasing, and the chart gains a third symbol. For elementary formulas this happens only for genuinely constant pieces; isolated zeros, the common case, do not interrupt monotonicity, as x + sin(x) demonstrates.
How does this connect to local maxima and minima?
Directly: an up-to-down switch is a local maximum, down-to-up a local minimum, no switch means neither. Reading extrema off the sign chart IS the first derivative test; this page computes the chart, and the companion test page phrases the conclusions per critical point.
Why does the dashed derivative curve help?
It turns the analysis into geometry: f climbs exactly where the dashed curve is above the axis, peaks exactly where the dashed curve crosses downward. Once you can read the pair, you can sketch f from f′ alone, a core exam skill and the reverse of differentiation's usual direction.
Does increasing imply positive derivative everywhere?
Not quite: increasing implies f′ ≥ 0 where the derivative exists, and zeros are permitted at isolated points (x³ at 0). The exact statement, monotone if and only if f′ keeps one weak sign, with strictness when zeros are isolated, is a Mean Value Theorem corollary.
What about functions with breaks, like 1/x?
Run the chart on each connected domain piece separately. 1/x decreases on (−∞, 0) and on (0, ∞), but joining the claims across the pole is false (compare f(−1) = −1 with f(1) = 1). Domain breaks are mandatory split points regardless of the derivative's sign.
How is monotonicity used to prove inequalities?
To show g(x) ≥ h(x) for x ≥ a, set f = g − h, check f(a) ≥ 0 and f′ ≥ 0 beyond a: a function starting non-negative and never decreasing stays non-negative. Enormous numbers of classical inequalities, eᵀ ≥ 1 + x among them, are two-line proofs in this pattern.