Key takeaways
- A growing-sequence game adds one item per round, so the failure point is not random: it arrives where the list first exceeds what you can hold as chunks rather than as items.
- Errors in immediate serial recall follow a bow-shaped curve. The first items are protected and the last items get a smaller lift, so the middle of the list is where a run usually comes apart.
- Jones and Oberauer report that forward recall typically shows strong primacy with only a little recency, while backward recall shows the reverse, so which end is protected depends on the order you reproduce the list in.
- Grouping a sequence into runs of three or four and giving each run its own rhythm converts a long list into a short one, which is the single most effective strategy on this format.
- Improvement on a sequence game is a better encoding habit, not a larger memory. The working-memory literature describes a central store of roughly three to five chunks that strategy works around rather than expands.
- IQ Test & Brain Training Games records a longest-sequence figure as a personal best, which is a record of your play in one mode and not a measurement of anything.
A growing-sequence game plays you a pattern, asks you to repeat it in order, then adds one item and does it again. That escalation is the whole design: it is a staircase procedure that climbs until it finds the length you cannot hold. What is less obvious is that the failure, when it comes, is usually not at the end of the list. It is in the middle — and that is a property of the task, not of you.
The format, stated plainly
Strip away the lights and sounds and this is immediate serial recall: a list is presented once, and you reproduce it in the original order straight away. Order matters, so producing the right items in the wrong arrangement is still a failure.
The growing part adds a staircase. Round three asks for three items, round seven asks for seven. That accumulation is the opposite of what an n-back task does, where the load stays flat and only the contents change. Because the list only ever grows by one, the game converges on your limit with reasonable precision, and because it starts easy it gives you several rounds of momentum before it gets interesting. That combination — self-calibrating difficulty plus a visible number at the end — is why the format has survived since the 1970s in one form or another.
In our own app, IQ Test & Brain Training Games, built by Reign Creative, this format appears as Simon Memory, described in the mode list as a twelve-round game where the pattern grows by one every round. Nearby modes vary the same idea: Color Sequence and Letter Sequence change the content, Order Recall changes the probe, and Backward Tap reverses the direction of reproduction. The app stores a longest-sequence figure among its personal bests, which is a record of your play in one mode rather than a measurement of anything.
Why the middle is where it breaks
Ask people to recall a list immediately and plot accuracy against position, and you do not get a flat line or a steady decline. You get a bow: relatively high at the start, dipping through the middle, lifting somewhat at the end.
The opening items are protected because they get rehearsed the most — they were in your head alone before anything else arrived, and every subsequent rehearsal pass includes them. The closing items are protected because they are the freshest. The interior positions get the least of both.
Jones and Oberauer, in a study of serial-position effects for items and relations in short-term memory, describe the standard pattern directly: forward recall typically shows primacy with only a little recency, whereas in backward recall the output order is the reverse of the input order and the resulting curve typically shows large recency and only a little primacy. In other words, which end is protected depends on the order you are asked to produce, and the disadvantaged region moves with it. They also report primacy and recency effects for relation memory — memory for what went next to what — with a more pronounced bow shape.
For a Simon-style run this has one clear practical consequence: when you feel the pattern slipping, it will almost always slip somewhere around positions four to six of an eight-item list, not at position eight. If you are diagnosing your own failures, look there first.
Direction matters more than people expect
A small but useful complication: forward and backward versions of these games are not the same task with a flipped switch.
In a review of forward and backward order recall, Donolato, Giofrè and Mammarella report overwhelming evidence for a distinction between forward and backward digit span — reversing a spoken list is genuinely harder, because it adds a transformation on top of the storage. But they also report no clear evidence for the same distinction in spatial span, and note that in visuospatial tasks the backward disadvantage shows up mainly in people with weaker spatial abilities.
So if you play a forward colour sequence and a backward tap version and find the gap between them small, that is a normal result for a spatial task rather than a sign that you are doing something clever. If you play forward and backward digit versions and find a large gap, that is also normal.
The strategy that actually moves the break point
There is one, and it is not "concentrate harder".
Group the sequence and give each group a rhythm. Take an eight-item colour run. Held as eight separate things it is far past the limit Nelson Cowan's review describes — a central store of roughly three to five meaningful chunks in young adults. Held as three groups, red-blue-green / yellow-red-red / blue-green, it is three things plus three small internal orders. The items have not changed. The number of things you are holding has.
Three rules make grouping work rather than backfire:
- Fix the group size early and keep it. Threes are the usual default, with a shorter final group. Re-deciding where the boundaries fall halfway through a run is a reliable way to corrupt a list you had already partly encoded.
- Use rhythm, not just position. Saying or tapping each group as a unit with a pause between groups gives the boundaries a physical marker, which is much easier to hold than an abstract decision about where group two ended.
- Rehearse groups, not items. When you go back over the list during the gap, run the three group-units rather than the eight items. Rehearsing item by item costs more and protects the interior less.
This is the same chunking mechanism described in our guide to memory techniques that actually work, applied to the one format where it pays fastest. It is also the reason that people who look impressive on these games are rarely holding more than you — they are holding fewer, larger units.
What improvement here does and does not mean
Expect your longest sequence to climb, then flatten. The climb is real: you are learning a specific encoding routine and applying it consistently, which is exactly the kind of gain practice reliably produces. The flattening is also real, because the underlying store is small and strategy works around it rather than expanding it.
What you should not conclude from a rising number is that your memory has improved in general. Near-transfer to similar tasks is comparatively well supported; broad transfer is disputed and weak, and we set the evidence out honestly in our guide to working memory and whether training changes it. A sequence game is practice and entertainment, and it is not a standardised assessment. IQ Test & Brain Training Games does include a memory test, which its listing says tracks recall against your own history, but the listing describes that result too as an estimate based on app activities, not a clinical or professional assessment.
Reading a sequence game as a design object
If the format interests you as a design rather than as a challenge, two things are worth noticing. First, the staircase does the difficulty tuning for free — the game never has to guess your level, because it walks up to it. That is unusual, and it is why the format tolerates such a wide range of players. Our piece on how quiz difficulty works covers the alternative approaches and their trade-offs.
Second, a growing sequence is one of the few formats where the failure is informative. Losing a run tells you your current chunked capacity on that content type at that reveal speed, which is a genuinely useful number to know about yourself even though it is not a measurement of anything psychological. Other formats and what each one demands are compared in memory games beyond matching pairs, and the rest of the span and encoding material sits in the memory and attention hub alongside the wider education and brain training category. Simon Memory and the other sequence modes are in IQ Test & Brain Training Games.
Good to know
Frequently asked questions
How does a Simon-style memory game work?
Why do I always lose the middle of the sequence?
What is the best strategy for getting further?
Does getting better at sequence games improve my memory?
Sources
- Serial-position effects for items and relations in short-term memory (opens in a new tab)Memory (Psychology Press), author copy hosted by the University of Zurich — accessed
- Differences in Verbal and Visuospatial Forward and Backward Order Recall: A Review of the Literature (opens in a new tab)Frontiers in Psychology, via PubMed Central (PMC5415597) — accessed
- The Magical Mystery Four: How is Working Memory Capacity Limited, and Why? (opens in a new tab)Current Directions in Psychological Science, via PubMed Central (PMC2864034) — accessed
About the author
Reign Creative LLC — The independent app studio that builds the apps covered on this site and publishes every article here. On Google Play it publishes as Reign Collective Apps.
We use AI tools to help research, draft and edit our articles. Each article lists the sources its factual claims rely on, and details about our apps come from their live Google Play listings. No article on this site has been reviewed by an independent expert unless the reviewer is named on it.
Spotted an error? Tell us — how corrections are handled is set out in our editorial policy.