Difference Threshold

Development of the Concept

The difference threshold, otherwise known as the just notable difference or the difference limen, is the smallest difference between two stimuli that can be consistently and accurately detected in experimental trials 50% of the time.

This concept describes the minimum amount by which a stimulus’s intensity must change to produce a noticeable difference. That difference is measured in the sensory experience of a participant taking part in an experiment.

Weber’s Law: The Original Study

Ernst Heinrich Weber (1795-1878) was a German physiologist, not a psychologist by training. He held a chair in physiology at Leipzig. There, his research on touch and the muscular sense raised a concrete question about how the senses work.

The relationship between these two is called Weber’s law.

The Weber Fraction: A Worked Example

In other words, Weber’s law says the difference threshold is in direct proportion to the original value of the stimulus. This proportion is the Weber fraction. Psychologists write it as ΔI/I = k, where ΔI is the just-noticeable increase and I is the starting intensity.

For example, a researcher gives an observer two spots of light, each with an intensity of 100 units. The observer increases one spot’s brightness until it looks just noticeably brighter than the other.

The observer needs the light to reach 110 units. The difference threshold here is ten units, so the Weber fraction is 0.1.

Using this fraction, the researcher can predict the difference threshold at other starting intensities. A 1,000-unit light needs to reach 1,100 units. A 10,000-unit light needs to reach 11,000 units, ten times the increase for ten times the light (Ross, 1995).

This law can be applied to any of the senses. The difference threshold can apply across brightness, loudness, mass, line length, and more.

The size of the Weber fraction itself varies across each sense, and from observer to observer. Still, the proportional relationship tends to hold (Ross, 1995).

Fechner and the Weber-Fechner Law

This proportional relationship between what people perceive and a physical quantity is the foundation of psychophysics.

Weber’s colleague at Leipzig, Gustav Fechner, founded this field in his 1860 book Elemente der Psychophysik (Fechner, 1860).

Fechner’s insight was simple. A series of just noticeable differences, stacked one on top of another, could be treated as a chain of subjectively equal steps.

Counting how many jnds separate a stimulus from the absolute threshold let Fechner assign a number to a sensation’s strength.

This extension is called the Weber-Fechner law.

It states that perceived intensity grows roughly logarithmically with physical intensity, so larger increases in a stimulus produce smaller and smaller increases in felt sensation.

Examples

Absolute vs. Difference Threshold

One concept that is often confused with the difference threshold is the absolute threshold.

While the difference threshold involves an observer’s ability to detect a difference in stimulation levels, the absolute threshold refers to the smallest detectable level of stimulation.

Hence the word absolute.

For example, the absolute threshold for sound is the quietest volume a person can just detect.

The difference threshold, by contrast, is the smallest change in volume that same person can notice once a sound is already playing.

The two concepts differ in another way, too. The absolute threshold is fixed for a given type of stimulus.

The difference threshold instead depends on the starting point: it changes depending on how loud, bright, or heavy the original stimulus already is.

Take light and sound as examples.

The absolute threshold for light is the dimmest level someone can detect in a room, and the absolute threshold for sound is the faintest noise they can hear.

Neither one changes.

The difference threshold works differently. For light, it is the smallest change in brightness someone can notice, whether the room starts off bright or dim.

For sound, it works the same way: the threshold stays proportionally the same, regardless of how loud or soft the original noise was.

Implications

The difference threshold has many implications for experimental psychology.

Namely, the difference threshold helps psychologists understand why people do or do not notice their own progress as they move through an experiment.

Picture a taste study. A researcher gives participants cups of ice cream with varying amounts of sugar, then has each participant rate how sweet each cup tastes.

If participants reliably rated one cup as sweeter than another, the sugar difference between the cups was above the just noticeable difference.

If the ratings showed no consistent difference, the sugar difference was too small to detect. It never crossed the difference threshold.

The same idea explains everyday experience. Someone may not notice a gradual change in their own weight, even as it happens over time, because each day’s change is too small to cross the threshold.

Marketing and advertising use this concept too. A company might weigh the difference threshold for brightness when designing product packaging. It wants the product to stand out on the shelf, but not so brightly that it becomes annoying.

By choosing a minimal discernable amount of brightness for their packaging, the company’s product can stand out from the display while minimizing the effects of visual irritation (Vojtko, 2014).

Real-World Applications of the Difference Threshold

The difference threshold is not just a laboratory curiosity. It defines the smallest change people will actually notice.

That makes it a practical constraint for anyone designing a product, running a clinic, or setting a price.

Marketing, Pricing and Shrinkflation

The clearest commercial example is shrinkflation: shrinking a product’s size while keeping its price the same, rather than raising the price outright.

If the size reduction stays below shoppers’ difference threshold, many will not consciously notice it.

Even so, they are now paying more per ounce.

Janssen and Kasinger (2026) studied a decade of grocery-store data.

They found that consumer demand barely responds to the resulting rise in the effective, per-ounce price.

Shoppers react far more strongly to an equivalent hike in the sticker price than to a same-sized cut in package contents.

The same logic works in reverse for pricing.

A firm that wants to raise a price quietly keeps the change below the threshold.

A discount only feels meaningful once the price cut exceeds it (Vojtko, 2014).

Engineering and Quality Control

Audio and lighting engineers must scale any adjustment to the current level, not apply it as a fixed step.

A small increase is obvious on a quiet amplifier, but the same increase is inaudible on a loud one.

This is one reason volume and brightness controls are often built on logarithmic rather than linear scales.

The same logic guides manufacturing.

The difference threshold sets the tolerance for two production runs of the same item, whether that is a paint colour, a scent, or a mechanical part’s fit.

Engineers define acceptable variation limits that stay safely under the threshold of customer complaint.

That keeps costs down, and customers never notice.

A tolerance set below the threshold is invisible to the people who use the product.

Clinical and Sensory Testing

In audiology, tests such as the Short Increment Sensitivity Index present a patient with a series of small volume increments.

The clinician checks whether the patient can detect them.

An unusually good ability to detect these tiny increments can signal cochlear hearing loss.

Vision testing uses the same logic.

A clinician presents light stimuli that step up in small increments to map the smallest brightness change a patient can detect at each point in their visual field.

Deviations from the expected pattern can flag early damage from conditions such as glaucoma.

The food industry relies on the same principle.

In the triangle test, panellists taste three samples, two identical and one different, and try to identify the odd one out.

If they cannot, a recipe change has stayed below the threshold and can be introduced without shoppers noticing.

Critical Evaluation of the Difference Threshold

Weber’s law has remained a working tool in psychology for close to two centuries.

It is not the last word on how sensation should be measured, though. Four issues are worth weighing.

• A Durable Regularity: one of psychology’s oldest surviving, replicable quantitative laws.
• Breaks Down at the Extremes: the proportional rule holds best in the middle of the intensity range, not at its limits.
• A Decision-Making Blind Spot: the classical model ignores how an observer’s own response bias shapes what they report noticing.
• An Unstable Weber Fraction: modern meta-analyses show the “constant” fraction for a sense varies far more than textbook tables suggest.

A Durable Regularity

Weber’s law has worked as a reliable tool for close to two centuries.

That is longer than almost any other finding in psychology.

It gave the young discipline of experimental psychology something rare at the time.

Here was a genuinely lawful, replicable, mathematically expressible regularity about the mind, not just a descriptive claim.

Few ideas from psychology’s first century have survived testing this well.

This continuity is well documented.

Weber’s proportional relationship still organises psychophysical research on sensory discrimination in essentially the same form today (Ross, 1995).

The jnd it defines gave psychophysics a repeatable, quantitative outcome measure at a time when the field had few others.

Fechner valued this so highly that he built the whole of psychophysics on top of it.

Breaks Down at the Extremes

Weber’s law holds best through the middle of an intensity range.

It fails near the absolute threshold, where discrimination becomes disproportionately difficult.

It also fails near the upper ceiling of a sense’s range, where receptors begin to saturate.

Loudness discrimination provides the clearest documented case.

Classical theory predicts the Weber fraction should stay constant as volume rises.

Instead, the fraction for pure-tone intensity discrimination actually shrinks (McGill & Goldberg, 1968).

Researchers call this the “near miss” to Weber’s law.

Hearing becomes proportionally more sensitive, not equally sensitive, as sounds get louder.

That means even one of the textbook example senses does not obey the law exactly across its full range.

The exception matters because it shows Weber’s law is an approximation, not a physical constant.

A Decision-Making Blind Spot

The classical model assumes discrimination is a purely sensory matter: either the system can register a difference, or it cannot.

Signal detection theory challenges this directly.

Sensory channels are inherently noisy.

So whether someone reports noticing a difference depends on more than their raw sensitivity.

It also depends on their decision criterion: how willing they are to say “yes, different” under uncertainty.

Two people can report different thresholds for an identical physical change.

So can the same person on two different days.

The cause is a shift in response bias, not a real change in sensitivity.

Signal detection theory does not replace Weber’s law so much as reinterpret it.

A measured jnd is not a hard physical boundary.

It is the point at which a signal reliably clears both the background noise and the observer’s own decision criterion.

Contemporary Research

A 2025 meta-analysis puts real numbers on how far this instability runs.

Worsley, Schroeder, and Dixit (2025) pooled effect sizes from published studies testing magnitude discrimination across diverse species and stimulus dimensions, including size, intensity and number.

On average, discrimination followed the pattern Weber’s law predicts.

But the strength of this effect varied widely from study to study.

None of the biological or methodological factors the authors examined explained that variation.

The gap remained unexplained.

A parallel meta-analysis of over 68 number-comparison studies reached a similar conclusion from a different angle (Guillaumé & Van Rinsveld, 2018).

Reported Weber-fraction values for the same underlying ability ranged from below 0.10 to above 0.60 in young adults.

The spread depended largely on which experimental method a study used, not on any real difference in ability.

The lesson repeats.

Together, these two meta-analyses support the classic finding that discrimination follows a Weber’s-law-like proportional pattern.

They also show that one fixed Weber fraction, quoted the way textbooks often do, understates how much the number depends on how, and in whom, it was measured.