How Long Does Short-Term Memory Last?
A phone number can disappear before you finish dialling it, yet another can remain as long as you keep repeating it. Short-term memory has no single countdown that applies in every situation.
The usual answer is that short-term memory lasts about 15 to 30 seconds without rehearsal. That range is useful as a basic introduction, but it can also be misleading. Information does not automatically vanish at the same moment every time, and the passage of time is not the only reason it becomes unavailable.
Retention depends on what you are remembering, how closely you attended to it, whether you continue refreshing it, and what other information enters your mind during the delay.
A practical way to see these limits is to try several kinds of immediate recall rather than relying on one number. Cognitive Train’s assessment measures verbal, numerical, visual, and spatial short-term memory in one session.
Where the 15-to-30-Second Answer Comes From
One of the best-known experiments was conducted by Lloyd and Margaret Peterson in 1959. Participants briefly saw a three-letter combination such as “KLP,” followed by a three-digit number. They then counted backward by threes to prevent themselves from silently repeating the letters.
In their short-term retention experiment, recall declined sharply as the delay increased from a few seconds to 18 seconds. The study became a foundation for the claim that unrehearsed short-term memories fade within roughly 20 seconds.
But the experiment did not simply leave participants sitting quietly while time passed. Counting backward filled the delay with another demanding mental activity. That prevented rehearsal, but it also created distraction and interference.
The famous result therefore shows that small amounts of unfamiliar verbal information can become difficult to retrieve very quickly when rehearsal is blocked and attention is occupied. It does not prove that every short-term memory has a built-in 18-second expiry time.
Short-Term Memory Is Not a Fixed Timer
If you repeat a phone number to yourself, it may remain available for considerably longer than 30 seconds. You are not demonstrating that short-term memory naturally lasts longer. You are actively refreshing the information before it becomes inaccessible.
The moment rehearsal stops, retention usually becomes less reliable. A conversation, notification, calculation, or new set of numbers can displace or compete with what you were trying to hold.
This is one reason the distinction between temporary storage and active mental processing matters. Cognitive Train’s guide to working memory versus short-term memory explains how simply holding information differs from holding it while performing another operation.
Does the Memory Fade, or Does Something Interfere?
The early experiments were often interpreted as evidence for decay: a memory trace supposedly weakens as time passes unless rehearsal renews it.
Later research complicated that explanation. In 1962, Geoffrey Keppel and Benton Underwood examined performance across repeated trials. Their study of proactive interference found that the steep loss was much smaller on participants’ first trial and became stronger after they had encountered several similar letter combinations.
Earlier material was interfering with newer material. Instead of each memory simply dissolving with time, similar items from previous trials were competing with the item participants were currently trying to recall.
A later series of experiments by Berman, Jonides, and Lewis attempted to separate the effects more directly. Their 2009 study of verbal short-term memory found a very small effect from the mere passage of time but a much larger effect from interference.
The safest conclusion is not that decay never occurs. It is that a clock alone does not explain most short-term forgetting. What fills the interval often matters more than the interval itself.
Why a Quiet Delay Is Different From a Busy One
Imagine looking at the code 583921 and waiting ten seconds before typing it. If the room is quiet and you keep the number in mind, recall may be easy.
Now imagine that someone asks you a question during those ten seconds. You listen, understand the question, prepare an answer, and then return to the code. The same amount of clock time has passed, but the mental conditions are completely different.
The new task captures attention and introduces competing verbal or numerical material. Even if the original code has not completely disappeared, its position among the competing information becomes harder to recover.
This is part of the broader explanation of why recently learned information becomes unavailable. Apparent memory loss can reflect weak encoding, interference, disrupted maintenance, or failed retrieval rather than one uniform fading process.
Capacity and Duration Are Different Limits
Short-term memory has at least two separate constraints: how much information can be held and how long it remains accessible.
You may remember four digits after a delay but fail immediately with twelve digits. In that case, capacity rather than duration caused the failure. The information exceeded what you could maintain as separate items.
Chunking changes this limit. The sequence 1-9-4-5 may require four separate units if it is random to you, but “1945” can function as one meaningful unit if you recognize the year. The number of visible characters has not changed, but the number of mental chunks has.
Cognitive Train’s article on working-memory capacity examines why modern estimates are often lower than the familiar “seven plus or minus two” figure once rehearsal and grouping strategies are controlled.
Try the Difference Between Immediate and Delayed Recall
Digit span is a simple way to experience how capacity, attention, and maintenance interact. Forward span mainly requires you to preserve a sequence in order. Backward span adds manipulation because you must hold the digits while mentally reversing them.
Your result is not a measurement of one universal short-term-memory timer. It reflects several processes working together: attention during presentation, the number of items, rehearsal, sequence organization, and accurate retrieval.
How to Keep Information Available Longer
Give it full attention. Information cannot remain available if it was never clearly encoded. Stop speaking, reading, or switching screens for the few seconds in which the information arrives.
Rehearse strategically. Repeat a code or instruction when you need it only briefly. For anything that matters beyond the next minute, repetition alone is not enough.
Reduce competing input. Avoid starting another verbal or numerical task while holding fragile information. Even a short interruption can create more interference than several quiet seconds.
Group related items. Turn separate elements into meaningful chunks. A date, phrase, pattern, or familiar category is usually easier to maintain than the same information as unrelated pieces.
Connect it to meaning. Information lasts beyond the immediate task when it becomes associated with knowledge already stored. Explain it, picture it, categorize it, or retrieve it after a delay rather than relying indefinitely on mental repetition.
So, How Long Does Short-Term Memory Last?
Without rehearsal and while attention is occupied, unfamiliar information may become difficult to recall within seconds. The common estimate of roughly 15 to 30 seconds describes this kind of fragile, unsupported retention, not a fixed biological deadline.
With active rehearsal, information can be kept available longer. With meaningful encoding and successful consolidation, it may stop being merely short term and become retrievable much later. With distraction or similar competing material, it can become inaccessible almost immediately.
The better question is therefore not only “How much time has passed?” It is also “What happened in the mind during that time?”
For a broader look at memory, attention, reasoning, and processing performance, explore Cognitive Train’s collection of free brain tests. You can also use the site’s wider brain training and cognitive-testing tools to compare how different forms of immediate recall behave under different demands.