A machine receives an input, doubles it and adds three. Its final output is 17. What was the input?
We can work backwards. Subtract three from 17 to obtain 14, then divide by two. The starting input was 7.
This is an elementary example of an important mathematical idea: some processes can be reversed, and the rule for reversing them can itself be a function.
Additional Mathematics formalises this through inverse functions. Students learn to distinguish a function from its inverse, to recognise when a reversal is possible, and to understand why some processes lose information that cannot be recovered uniquely.
That is a compelling reason to study A-Math. The subject does not only ask, “What output will this input produce?” It also asks, “Which input could have produced this output, and is that input uniquely determined?”
This way of reasoning supports algebra, graph interpretation and more advanced quantitative study. It also develops caution about a common assumption: that every mathematical process can be undone in exactly one way.
Start by understanding the forward rule
Consider f(x) = 2x + 3.
The function doubles its input and adds three.
When x = 7, f(7) = 17.
When x = 0, f(0) = 3.
When x = -2, f(-2) = -1.
Over the real numbers, every real input gives exactly one real output.
The function is also one-to-one. If two inputs produce the same output, those inputs must be equal.
We can demonstrate that mathematically.
Suppose f(a) = f(b).
Then 2a + 3 = 2b + 3.
Subtract three from both sides and divide by two.
We obtain a = b.
Therefore distinct inputs cannot share the same output.
That property is what allows the function to be reversed uniquely over the real numbers.
Derive an inverse instead of memorising one
Let y = 2x + 3.
We want to find x in terms of y.
Subtract three:
y – 3 = 2x.
Divide by two:
x = (y – 3)/2.
Now exchange the variable labels to express the inverse function in familiar notation:
f⁻¹(x) = (x – 3)/2.
The symbol f⁻¹ names the inverse function.
It does not mean the reciprocal 1/f(x).
This distinction matters because the small superscript -1 has different interpretations in different mathematical contexts.
A reciprocal of 2x + 3 is 1/(2x + 3), which is not the inverse function of 2x + 3.
The inverse function reverses the original mapping.
Check the reversal in both directions
A useful way to verify an inverse is to compose the functions.
First apply f and then f⁻¹:
f⁻¹(f(x)) = ((2x + 3) – 3)/2 = x.
Now reverse the order:
f(f⁻¹(x)) = 2((x – 3)/2) + 3 = x.
Both compositions return the original input.
This is exactly what we expect from two mutually inverse rules when the input belongs to the appropriate domains.
The identity f⁻¹(f(x)) = x does not mean that inverse functions perform no operation. It means the second operation undoes the effect of the first.
In an actual worked question, checking both compositions is a strong way to detect a sign error in the proposed inverse.
Graphs show the inverse relationship visually
A function maps an input x to an output y.
Its inverse exchanges those roles.
If the point (7, 17) lies on y = f(x), then the point (17, 7) lies on y = f⁻¹(x).
These coordinates are reflections of one another across the line y = x.
The graph of an invertible function and the graph of its inverse are reflected in that diagonal line.
For the example f(x) = 2x + 3, the line has gradient 2.
The inverse f⁻¹(x) = (x – 3)/2 has gradient 1/2.
These are not perpendicular lines. Their gradients are reciprocals because reflection across y = x exchanges horizontal and vertical changes.
A student who confuses inverse reflection with perpendicularity may reach an incorrect conclusion.
The diagram is helpful, but the algebra determines the precise relationship.
Not every function has an inverse function over its whole domain
Now consider g(x) = x² over all real numbers.
We know g(3) = 9.
But g(-3) = 9 too.
Two different inputs produce the same output.
If we ask, “What input produced 9?”, the answer could be 3 or -3.
There is no unique answer.
Therefore x² on the entire real-number domain does not have an inverse function.
This is a subtle but important distinction. The equation y = x² can be rearranged to x = ±√y, but those two branches do not together define a single-valued inverse function for the original mapping over the whole real domain.
Additional Mathematics teaches students to check whether the requested reversal actually exists before writing an inverse formula.
Restricting the domain can make reversal possible
Return to g(x) = x².
Suppose we restrict the original input to x ≥ 0.
Now only the non-negative branch remains. Each output y ≥ 0 comes from exactly one allowed input.
The inverse is g⁻¹(x) = √x for x ≥ 0.
If instead we restrict the original input to x ≤ 0, the inverse becomes g⁻¹(x) = -√x for x ≥ 0.
The algebraic expression x² has not changed.
The chosen domain has changed.
That change determines which inverse is valid.
This is why a function should be understood together with its domain rather than treated as a formula floating independently of its allowed inputs.
A worked restricted quadratic
Consider f(x) = (x – 2)² + 1 with domain x ≥ 2.
This is an upward-opening parabola whose turning point is (2, 1).
Because the chosen domain includes only the right-hand branch, the function increases as x increases from 2.
Its range is y ≥ 1.
To find the inverse, start with:
y = (x – 2)² + 1.
Subtract one:
y – 1 = (x – 2)².
Since x ≥ 2, we know x – 2 ≥ 0.
Therefore:
x – 2 = √(y – 1).
So:
x = 2 + √(y – 1).
The inverse function is f⁻¹(x) = 2 + √(x – 1), defined for x ≥ 1.
Notice how the original domain restriction determined the positive square-root branch.
If we had chosen x ≤ 2 instead, the inverse expression would have required the negative branch.
Domain and range exchange their roles
In the restricted quadratic example, the original function accepts x ≥ 2.
Its outputs satisfy y ≥ 1.
The inverse therefore accepts inputs x ≥ 1, because its input values are the original outputs.
Its outputs satisfy y ≥ 2, because it recovers the original input values.
In short:
- The domain of the original function becomes the range of the inverse.
- The range of the original function becomes the domain of the inverse.
This relationship helps students test an inverse without doing a long calculation.
If a proposed inverse has a domain that contradicts the range of the original function, something is wrong.
It also helps explain why the reflection across y = x exchanges horizontal and vertical features of the graph.
A check with a particular input
For f(x) = (x – 2)² + 1, x ≥ 2, choose x = 5.
Then f(5) = 3² + 1 = 10.
Now apply the inverse to 10:
f⁻¹(10) = 2 + √9 = 5.
The original input is recovered.
Now choose x = 2.
The original function gives f(2) = 1.
The inverse returns f⁻¹(1) = 2.
This checks the boundary values.
A careful student should test at least one ordinary value and, where useful, an endpoint after deriving an inverse.
The tests do not by themselves prove the formula correct for every input, but they can expose simple algebra or branch-selection mistakes.
Why monotonic behaviour makes inversion easier
A function is strictly increasing on an interval when larger inputs produce larger outputs.
A strictly decreasing function behaves in the reverse way: larger inputs produce smaller outputs.
Either kind of strict monotonic behaviour guarantees that different inputs within that interval have different outputs.
That makes the function one-to-one on the interval.
For a differentiable function, derivative signs can help us determine where it is increasing or decreasing.
Consider f(x) = x³ + x.
Differentiate:
f'(x) = 3x² + 1.
Because 3x² + 1 is always positive, f increases strictly throughout the real-number domain.
Therefore it is one-to-one.
It has an inverse function on its range, even though writing its inverse as a simple elementary expression is not as convenient as for a linear function.
The existence of an inverse and the ease of calculating its formula are two different matters.
This is a valuable distinction for mathematically mature students.
A cubic whose reversal needs care
Consider h(x) = x³ – 3x.
Differentiate:
h'(x) = 3x² – 3.
This derivative is positive for x < -1 and x > 1, but negative for -1 < x < 1.
The function increases, then decreases, then increases.
It is not one-to-one over the whole real-number domain.
For example:
h(0) = 0.
h(√3) = 3√3 – 3√3 = 0.
h(-√3) = -3√3 + 3√3 = 0.
Three distinct inputs produce the same output.
A whole-domain inverse function therefore does not exist.
However, a suitable interval restriction can select a branch on which the function is strictly monotonic.
The important task is to identify the valid interval before attempting reversal.
The horizontal-line test
Graphically, a function is one-to-one if every horizontal line intersects its graph at most once.
Why?
A horizontal line y = c collects all inputs producing the output c.
If it meets the function graph twice, the same output comes from two different inputs.
That prevents unique reversal.
The graph of y = x² over all real x fails this test: the line y = 9 meets it at x = -3 and x = 3.
The restricted right-hand branch with x ≥ 0 passes the test.
The test is a visual representation of the one-to-one condition.
It does not replace careful algebra when the graph is rough or its viewing window hides important behaviour, but it provides an excellent conceptual check.
An inverse of a rational function
Consider:
f(x) = (2x – 1)/(x + 3), with x ≠ -3.
The denominator restriction is part of the original function.
We want its inverse.
Begin with:
y = (2x – 1)/(x + 3).
Multiply both sides by x + 3:
y(x + 3) = 2x – 1.
Expand:
xy + 3y = 2x – 1.
Collect terms containing x:
xy – 2x = -1 – 3y.
Factorise:
x(y – 2) = -(1 + 3y).
Therefore:
x = (1 + 3y)/(2 – y).
The inverse is:
f⁻¹(x) = (1 + 3x)/(2 – x), with x ≠ 2.
The restriction x ≠ 2 on the inverse is essential because its denominator cannot be zero.
Verify the rational inverse
First check a simple original input, x = 0.
Then:
f(0) = (-1)/3 = -1/3.
Apply the proposed inverse:
f⁻¹(-1/3) = (1 + 3(-1/3))/(2 – (-1/3)).
The numerator is 0. The denominator is non-zero.
So the result is 0.
The original input is recovered.
Now ask why the original function’s output can never equal 2.
If (2x – 1)/(x + 3) = 2, then multiplying gives 2x – 1 = 2x + 6.
That would require -1 = 6, which is impossible.
Therefore 2 is excluded from the original range.
That explains why 2 is excluded from the inverse domain.
This is a clear example of how algebraic restrictions and invertibility fit together.
Why an inverse is not the same as a reciprocal
This is such a common source of confusion that it deserves a separate comparison.
Let f(x) = 2x + 3.
The inverse function is:
f⁻¹(x) = (x – 3)/2.
The reciprocal of the function value is:
1/f(x) = 1/(2x + 3).
At x = 5, the inverse gives (5 – 3)/2 = 1.
The reciprocal gives 1/13.
These are not remotely the same result.
The symbol f⁻¹ refers to an inverse mapping in this context; it does not describe taking the reciprocal of the output.
Understanding notation prevents an avoidable mistake in many function questions.
Model a process that can be reversed
Suppose a hypothetical process estimates total cost C from the number x of units:
C(x) = 100 + 12x.
The number 100 represents a fixed component and 12 is the additional cost per unit in the model.
If x = 5, then C = 160.
Now suppose we know the total cost and want to recover the implied number of units.
Rearrange:
x = (C – 100)/12.
The inverse rule is C⁻¹(y) = (y – 100)/12, if we use C as the name of the forward function.
But the practical domain matters.
If x represents non-negative quantities, the model’s cost outputs are at least 100.
If x represents a whole number of units, valid outputs correspond to integer values of x.
For example, a cost of 166 would imply 5.5 units, which may not be permitted in a discrete-unit application.
The algebraic inverse is a tool. The interpretation still depends on the model.
Why reverse problems require a clear target
Many school word problems describe a forward process and ask students to work backwards.
A quantity is transformed. An amount is adjusted. A length is scaled. A value is converted.
The inverse method can help, but only when the underlying relation is appropriately one-to-one over the intended domain.
If two original quantities could produce the same observed result, more information is needed to recover one uniquely.
For instance, the squared distance from a point on a number line to the origin does not reveal whether the point was to the left or right.
An observed squared distance of 9 could come from -3 or 3.
Knowing that the position was non-negative removes the ambiguity.
This example shows why inverse questions are also questions about information.
A process that merges distinct inputs into one output cannot necessarily be reversed without extra conditions.
A two-stage process and its inverse
Let f(x) = 2x + 3 and g(x) = x – 5.
The composite g(f(x)) applies f first and then g.
Calculate:
g(f(x)) = (2x + 3) – 5 = 2x – 2.
To reverse the full process, add 2 and divide by 2.
Thus the inverse composite is (x + 2)/2.
Now consider reversing the stages individually.
The inverse of g(x) = x – 5 is g⁻¹(x) = x + 5.
The inverse of f(x) = 2x + 3 is f⁻¹(x) = (x – 3)/2.
To undo g(f(x)), first undo the last operation g, then undo f.
So:
f⁻¹(g⁻¹(x)) = ((x + 5) – 3)/2 = (x + 2)/2.
The results agree.
This is a general principle: when a sequence of invertible functions is reversed, the inverse operations are applied in the opposite order.
The idea becomes intuitive when students think of unfastening layers in the reverse order to which they were applied.
A case where reversing a composite is not unique
Let f(x) = x + 2 and g(x) = x² over all real inputs.
The composite g(f(x)) = (x + 2)².
Suppose the output is 25.
Then x + 2 = 5 or x + 2 = -5.
The possible inputs are x = 3 or x = -7.
The composite is not one-to-one over all real x.
Restricting the input to x ≥ -2 selects the branch x + 2 ≥ 0 and allows a unique inverse.
Without the restriction, the observed output does not identify which original input was used.
This example connects composite functions, quadratic equations and domain restrictions into a single problem.
When the graph of a function meets its inverse
The graphs of an inverse pair are reflections across y = x.
Sometimes they intersect on that line, although particular inverse graph pairs can also have intersections away from it in some settings.
For a simple linear example, take f(x) = 2x + 3.
A point fixed by f satisfies f(x) = x.
Solve:
2x + 3 = x.
Therefore x = -3.
The point (-3, -3) lies on both the function and its inverse.
The fixed-point interpretation is useful because input and output are equal there.
It is an enrichment connection rather than a rule to replace ordinary inverse-function methods.
Students should still construct and interpret both graphs carefully instead of assuming that every intersection must sit on y = x for all possible inverse pairs.
Why solving an inverse problem often begins with an equation
Suppose a question tells us the final output y and asks for the original input x.
An important first step is to write:
y = f(x).
Then solve for x.
If a unique solution exists for every y in the stated range, we can express the result as the inverse function.
This method works for many linear and suitably restricted nonlinear functions.
However, we must not automatically regard any algebraic rearrangement as a valid inverse function.
If solving for x gives two branches, the original function may not be one-to-one.
If a denominator becomes zero, the inverse has a restricted domain.
If a square root appears, the radicand must be non-negative.
Mathematical validity comes before neat notation.
The role of inverse functions in logarithms
An exponential function with a suitable positive base is one-to-one.
For example, f(x) = 2^x increases strictly for all real x.
Its outputs are positive real numbers.
The inverse function is logarithm base 2:
f⁻¹(x) = log₂x, with x > 0.
This explains why a logarithm can recover an exponent.
If 2^x = 32, the inverse tells us x = log₂32 = 5.
The relationship is not a separate memorised trick. It is an instance of inverting a function.
Students who understand inverse functions often find the purpose of logarithms clearer.
They also understand why taking a real logarithm requires a positive argument.
The valid outputs of the exponential become the valid inputs of the logarithm.
Why trigonometric inverses require care
The sine function repeats its values over different angles.
For example, sin 30° = 1/2 and sin 150° = 1/2.
So sine over all real angles is not one-to-one.
An inverse-sine function is defined by selecting a principal interval on which sine can be reversed.
That principal-value convention explains why a calculator returns one angle for inverse sine even when a trigonometric equation has several valid solutions in a larger specified interval.
Students must distinguish using an inverse trigonometric key from solving for every angle satisfying the original equation.
The inverse gives a principal value; periodicity and symmetry may supply additional solutions.
This is another case in which a domain restriction makes a useful inverse possible.
A simple independent verification routine
When a student derives an inverse function, they can follow a dependable checklist.
First, write the original function and its domain.
Second, determine whether it is one-to-one on that domain.
Third, set y equal to the original expression and rearrange for x.
Fourth, exchange variable labels to write the inverse rule.
Fifth, identify the inverse domain using the original range.
Sixth, check a valid numerical input and, if practical, verify the compositions.
Seventh, compare the original and inverse graphs or their known features.
This process is not busywork. Each step addresses a particular kind of error.
Students who follow it gradually learn to detect an invalid inverse before the final answer is submitted.
Common mistakes and targeted corrections
Mistake one: reversing without checking one-to-one behaviour
The student tries to define an inverse of x² on all real numbers.
The remedy is to compare inputs 3 and -3 and explain why the same output prevents unique reversal.
Mistake two: choosing the wrong square-root branch
The original function is a restricted quadratic, but the student selects a branch inconsistent with the stated domain.
The remedy is to inspect the domain before solving for the inverse.
Mistake three: confusing inverse with reciprocal
The student writes 1/(2x + 3) as the inverse of 2x + 3.
The remedy is to check whether composing the claimed inverse with the original returns x.
Mistake four: missing an excluded input
A rational inverse contains a denominator, but the student omits its restriction.
The remedy is to track the original range and the inverse domain.
Mistake five: reversing a multistage process in the wrong order
The student undoes the first step before undoing the last.
The remedy is to draw the forward arrows and reverse them one by one.
These corrections focus on the mathematical cause, not merely the final wrong expression.
A five-session home practice plan
Session one: linear inverse functions
Begin with f(x) = 2x + 3 and two other linear functions with non-zero gradients.
Derive each inverse and verify it through composition.
Session two: inverse graphs
Draw a function, the line y = x and its inverse.
Plot paired points such as (7, 17) and (17, 7).
Explain what reflection means for the input and output coordinates.
Session three: restricted quadratics
Work with f(x) = (x – 2)² + 1 on x ≥ 2 and x ≤ 2 separately.
Find the correct inverse for each branch and explain the difference.
Session four: domains and rational functions
Use a rational function with a denominator restriction.
Derive its inverse and identify any excluded value in the inverse domain.
Check the result with a numerical substitution.
Session five: composite reversals
Construct a two-stage process using simple invertible functions.
Find the full composite and then reverse it.
Check why the inverse stages appear in reverse order.
The goal is to make the reasoning transferable rather than dependent on one remembered formula.
Singapore’s Additional Mathematics context
The 2027 Singapore-Cambridge Secondary Education Certificate G3 Additional Mathematics syllabus, K341, includes functions, their inverses, composite functions and related domains and ranges.
Students need to read the official syllabus for the precise assessment scope and any specifications about methods.
The SEAB 2027 G3 syllabus directory identifies the current subject and links to its official document.
These worked examples are intended to explain the meaning of mathematical reversibility. They do not predict examination questions or guarantee how every school sequences the topic.
The lasting skill is learning to check whether a procedure can be reversed under its stated conditions.
How parents can support the learning
Parents can ask simple, revealing questions.
If the output is known, can the child explain how to recover the input?
Could two different inputs produce that same output?
If the function involves squaring, which input branch has been chosen?
Does the proposed inverse work when tested with an example?
Is an excluded value stated?
A student who explains these issues clearly is developing stronger reasoning than one who writes an answer without understanding why it is unique.
Parents should also notice whether mistakes arise from algebraic manipulation or from misunderstood function concepts. They are different learning problems.
How a small-group tutor can help
In a three-student Additional Mathematics class, one student may understand reversibility but struggle to rearrange a rational equation.
Another may be algebraically fluent but forget domain restrictions.
A third may know the formula yet confuse inverse with reciprocal notation.
At eduKateSG, the useful teaching response is to identify which step is unstable, explain the underlying idea and then test it with a changed example.
Secure lower-secondary algebra makes a major difference. The immutable Secondary 1 Mathematics tutor Clementi small-groups tutorials reference illustrates the importance of clear equation handling and independent working before advanced function questions arrive.
Students already progressing well with school teaching and their own corrections may not need tuition. Where a recurring difficulty persists, the next lesson should address the specific cause rather than assign unstructured repetition.
Frequently asked questions
Does every function have an inverse?
Not as a function on its entire original domain. A function must be one-to-one, relative to the relevant domain and range, for a unique inverse function to exist.
Why is the inverse of x² not simply √x?
The unrestricted real function x² has two inputs producing many positive outputs. Restricting the original domain to x ≥ 0 produces the inverse √x.
Is f⁻¹(x) the same as 1/f(x)?
No. The first refers to the inverse mapping, while the second is the reciprocal of the function value.
How do I know whether an inverse is correct?
Compose the original and proposed inverse, checking valid domains, and verify that the result returns the original input.
Why do domain and range swap?
The outputs of the forward function become the inputs of its inverse, while the original inputs become the inverse outputs.
Why must inverse operations be applied in reverse order?
The last step of a forward process must be undone first, otherwise an earlier stage may receive the wrong value.
Do inverse functions matter outside Mathematics?
They offer a useful abstract way to reason about recovering inputs from known outcomes. Real applications also require appropriate assumptions, units and evidence that the process is uniquely reversible.
The deeper lesson about information
A forward calculation may be easy even when reversing it is impossible.
Squaring 3 and -3 produces the same result. Knowing the square alone cannot tell us which original number was used.
A one-to-one rule preserves enough information for unique reversal. A many-to-one rule does not, unless the permitted input set is restricted.
Additional Mathematics makes that distinction visible through algebra, graphs and function notation.
Students learn to ask not only whether a calculation can be carried out, but whether the result contains enough information to recover where it came from.
That is a valuable reason to study inverse functions—and a useful habit of reasoning to carry into later education.
Explore the eduKateSG Additional Mathematics project library and the companion Why Study Additional Mathematics | Build Multi-Step Thinking with Composite Functions for the forward-process view. For checking conditions, see Why Study Additional Mathematics | Know When an Algebraic Answer Is Invalid.