011 min
The problem Common Region solves
Picture a recipe grid on a tablet. Every tile has a title, a byline and a star rating, and the tiles sit in rows. Some titles fit on one line and some need three. Presumably the layout leaves extra room under every title in case the title wraps.
The Food Network tablet app had exactly this problem, as the Nielsen Norman Group reported. Large gaps around each recipe in an earlier release of the Food Network tablet app made it hard to tell which byline and rating went with which recipe. The data was correct, because each rating was attached to the right recipe. The screen did not show it.
A later version of the app put each recipe's content inside a card. A border now marked where one recipe ended and the next began, and the reader no longer had to guess.
Designers call this fix the Law of Common Region. A boundary around a set of items makes people see those items as one group, and a boundary can outweigh the spacing around them. The next sections explain why that works, where the idea came from, and where it stops helping.
021 min
What the Law of Common Region says
The Law of Common Region explains why the border fixed the recipe grid when the spacing could not. When several items sit inside one enclosed area, people read them as a group. The enclosed area is the common region. The items do not need to look alike, and they do not need to sit close together. Sharing the area is enough.
Two older grouping cues make the effect easier to see. Proximity is the cue that items placed near each other look related. Similarity is the cue that items with the same shape or colour look related. A boundary can outweigh both. Take four circles in a row with equal gaps. People see four separate circles. Draw one outline around the middle two and people see a pair, although the gaps have not changed.
A region can be made in three common ways:
- A border line around the items.
- A background fill behind the items.
- A shadow or raised edge that makes the group look like its own surface.
All three share one cue: an enclosed area. Which one to use is a styling choice.
032 min
Where Common Region comes from
Gestalt psychologists listed the first grouping cues, including proximity and similarity, in the first half of the twentieth century. Their approach is based on the claim that the mind organises what it sees into simple, whole patterns, a tendency called the Law of Prägnanz. Common region was added much later.
Palmer's 1992 paper in Cognitive Psychology proposed grouping by common region and showed that it can overcome proximity and similarity. Palmer was a psychologist at the University of California, Berkeley. He built displays similar to the original ones from the Gestalt period and gave some of the elements a shared region. The shared region changed which elements people saw as a group.
Palmer also argued that region is a principle of its own and not a side effect of something older. Palmer argued that common region cannot be explained by proximity, closure or any earlier known factor.
The date that gets repeated wrongly
Several design articles date the principle to 1999 and credit Palmer and Rock. The 1992 paper has one author. Palmer and Rock did publish together in 1994, but their paper covers a different principle. That paper describes a connected area of the same colour or texture that is seen as one unit. The principle is called uniform connectedness. It is related to common region, but it is not the same idea and it is not from the same year.
042 min
How Common Region behaves in experiments
Palmer's paper and later experiments add details that matter to anyone who draws boxes in an interface. Four of them follow, each in its own part so a reader who wants only the basics can skip them.
Region depends on depth
Palmer found that grouping by region depends on perceived depth and is dominated by the smallest background area. In plain terms, an outline works as a region when it looks like the area behind its items. That is why a card with a fill or a soft shadow reads as a surface, while a few stray lines on a page do not.
The smallest region decides
When one region sits inside another, the items in the inner region group with each other first. In an interface, the contents of a card belong together before the cards belong together. The cards then group inside the panel that holds them. Each group can be handled as one chunk of information, which lowers the effort of reading the screen.
Boundaries and spacing add up
In 2011, Dolores Luna and Pedro Montoro tested common region together with proximity and similarity, and found that cues that agree produce stronger grouping than either cue alone. When the cues disagreed, grouping was weaker than with either cue alone. A card therefore works best when its border and its spacing point the same way, with more space between cards than inside a card. The experiments used simple patterns, so this is a design inference and not a measured result for interfaces.
Boundaries are noticed fast
In a 2016 study, Luna and colleagues found that people responded faster to groups formed by a boundary than to groups formed by shape similarity. The displays lasted 50, 150 or 350 milliseconds, so the test measured first-glance perception and not careful reading. That fits how cards work in practice: a reader sees which items go together before reading any of them.
052 min
How Common Region shows up in real interfaces
The Food Network cards are the usual case of Common Region, and the Microsoft Windows print dialog shows a second use. The dialog contains the numeral 1 in two places, one in the page range and one in the number of copies. Borders sort the options into three boxes: which printer, which pages and how many copies. In the Microsoft Windows print dialog, borders keep the page range apart from the number of copies, so the same numeral reads as a page number in one box and a copy count in the other.
The same cue appears in many other patterns. Slack's desktop app uses one dark background for its controls, which keeps them apart from the content.
| Pattern | What forms the region | What the reader concludes |
|---|---|---|
| Card | A border or fill around a tile | The photo, title and rating belong to one item |
| Feed post on a social site | A box around each post | The image, caption and comments belong to one post |
| Image and caption | A shared boundary | The caption describes this image |
| Accordion panel | One background for the label and its content | This label opens this content |
| Table with striped rows | An alternating row colour | The cells in a row describe one thing |
| Page header or app sidebar | A fill behind the controls | These controls belong to the interface and not to the content |
| Newspaper or magazine layout | A box or column around a story | The story and its picture go together |
Accordion panels are a common form of progressive disclosure, and the shared background is what lets a reader link each label to the content it opens. A sticky header with its own background colour also stays separate from the content that scrolls beneath it.
061 min
Applying Common Region
Knowing the effect leads to a short list of design moves.
Start with spacing.
Put related items close together and unrelated items far apart. Add a boundary only where spacing cannot do the job, for example when a group holds different kinds of content such as a photo, text and a button.
Choose the lightest cue.
A pale background fill often groups items without adding lines. Add a border when the fill is too close to the page colour to see, and add a shadow only when the group should look like a separate surface.
Match the boundary to the group.
Every item that belongs is inside it, and nothing that does not belong is.
Keep the gap between regions larger than the gap inside a region.
Then spacing and boundary agree.
Check nested regions.
The innermost box decides the grouping, so it should hold the items that are most closely related.
Test with a colleague.
Ask them to point at the items that belong together. Where they point differently from what you meant, the cues disagree.
Regions also shape the visual hierarchy of a screen, because each box becomes a unit that readers rank against the other units. On a small screen, where gaps are tight, a fill or border can group items where spacing cannot.
072 min
When Common Region goes wrong
Each of these moves can go too far, and the damage is easy to see once you know where to look.
Too many boxes
The Nielsen Norman Group judged the boxes around each filter group on the mobile Backcountry.com site unnecessary, because proximity already showed the grouping. Every extra box adds an edge for the reader to process and takes space away from content. Before adding a border, remove it in a mockup and check whether the grouping is still clear.
False floors
Full-width bands of colour cause a different problem. The Nielsen Norman Group calls the result a false floor, which is a point on the page that looks like the bottom but is not. Full-width blocks with their own border or background colour can make readers stop scrolling, because the colour change looks like the end of the page. A narrower band, or content that visibly continues past the edge of the band, removes the stopping point.
A weak boundary against strong spacing
Montoro and Luna reported in 2015 that how two cues combine depends on how strong each cue is alone. In their results, with one strong cue and one weak cue, agreement added little over the strong cue, and disagreement left grouping about as weak as the weak cue alone. A faint outline around items that wide gaps keep apart is a weak cue against a strong one, so the grouping stays unclear. Make the outline clear or close the gaps. Both studies used simple dot patterns and not interfaces, so treat this as a warning sign and test it on real screens.
081 min
Common Region and accessibility
Everything above assumes the reader can see the box. A screen reader announces items one after another and does not announce that a card begins or ends.
WCAG success criterion 1.3.1 names a different background colour as one way sighted users see that items are related. The criterion is called Info and Relationships, at Level A in WCAG 2.2. It requires that relationships shown by presentation can be worked out by software or are given in text.
In practice, give each region a matching structure in the code. Use a fieldset with a legend for related form controls, and use headings or lists so that each card has a clear start and end. People who replace a page's colours with their own style sheet also lose backgrounds that formed regions, and the structure in the code still carries the grouping for them.
091 min
Common Region vs. nearby concepts
Common Region is easy to mix up with the grouping cues around it. The question that separates them is whether an enclosure, drawn or implied, groups the items.
| Concept | What groups the items | Needs an enclosure? |
|---|---|---|
| This pageCommon Region | ||
| DesignProximity | ||
| Not in the library yetSimilarity | ||
| DesignUniform connectedness |
A simple test follows from this. Remove every border and fill from a layout. If the grouping survives, spacing alone grouped the items, and Common Region did not.
?4 questions
Questions people ask
Does Common Region need a visible border?
Is a card the same thing as a common region?
How many boxes are too many on one page?
Is a region the same as a container in code?
§8 sources
Sources
Show all 8 sourcesHide sources
Palmer, S. E. (1992). Common region: a new principle of perceptual grouping. Cognitive Psychology, 24(3), 436-447.
Palmer, S. E., & Rock, I. (1994). Rethinking perceptual organization: the role of uniform connectedness. Psychonomic Bulletin & Review.
Luna, D., & Montoro, P. R. (2011). Interactions between intrinsic principles of similarity and proximity and extrinsic principle of common region in visual perception. Perception.
Montoro, P. R., & Luna, D. (2015). Does the relative strength of grouping principles modulate the interactions between them? The Spanish Journal of Psychology, 18, E33.
Luna, D., Villalba-Garcia, C., Montoro, P. R., & Hinojosa, J. A. (2016). Dominance dynamics of competition between intrinsic and extrinsic grouping cues. Acta Psychologica, 170, 146-154.
Nielsen Norman Group. The principle of common region: containers create groupings.
W3C. Understanding SC 1.3.1: Info and Relationships (Level A), WCAG 2.2.
Laws of UX. Law of Common Region, a short summary for designers.





