Perceptual Organisation

Perceptual organisation = the process through which the brain organises visual information into meaningful shapes and forms.

Gestalt Principles

Gestalt principles are automatic grouping cues/rules the brain uses to organise visual elements into meaningful units.

They’re basically the visual system’s rules of thumb for deciding what belongs together.

Proximity

Things close together are perceived as belonging together.

Example:

● ● ● ● ● ●

You naturally see three pairs rather than six completely independent dots.

Simple rule

Close = probably belongs together.

Similarity

Things that look similar are grouped together.

Similarity can involve:

• colour
• shape
• size

Example: people wearing the same uniform may automatically be perceived as belonging to the same group.

Simple rule

Looks alike = probably belongs together.

Symmetry

Symmetrical or balanced elements tend to be grouped into a single unified shape.

Symmetry helps us recognise many natural and human-made objects.

Good Continuation

Lines and patterns are perceived as continuing smoothly.

If one line disappears behind an object and appears again on the other side, we tend to perceive it as one continuous line, rather than two unrelated lines.

Simple rule

The brain prefers smooth continuation.

Closure

The brain fills in missing information to perceive a complete shape.

For example, a circle with a small section missing is usually still perceived as a circle.

This helps us recognise:

• incomplete objects
• partially hidden objects
• smudged letters

Simple rule

Missing bit? Brain fills the gap.

Common Fate

Elements moving together are perceived as belonging together.

Example:

A group of birds flying in the same direction is perceived as a flock.

Simple rule

Moves together = belongs together.

Figure-Ground Organisation

Figure-ground organisation = separating an object from its background.

• Figure = the thing you’re focusing on
• Ground = the background

This is necessary because the visual system needs to determine:

“Which part of this image is the actual object?”

Rubin Vase

The classic Rubin Vase can be perceived as either:

a vase

OR

two faces

depending on which region is treated as the figure.

This demonstrates that the same visual information can be organised in different ways.

The physical stimulus hasn’t changed — the perceptual organisation has.

Gestalt Principles Summary

Principle	Brain assumes…
Proximity	Close things belong together
Similarity	Similar things belong together
Symmetry	Symmetrical elements form a unit
Good continuation	Lines continue smoothly
Closure	Missing information should be filled in
Common fate	Things moving together belong together
Figure-ground	Some regions are objects; others are background

Gestalt Cues Are Useful, But Not Perfect

Gestalt principles allow the brain to organise complicated scenes quickly and automatically.

But because they’re perceptual shortcuts, ambiguous stimuli can sometimes be organised in more than one way.

This helps explain:

• ambiguous images
• visual illusions
• confusing layouts/designs

The problem is proximity: the visual system can’t easily determine which label belongs with which box.

Image-Based Segmentation

Gestalt grouping isn’t the only thing happening during shape representation.

The visual system also needs to divide the scene into separate meaningful regions.

Segmentation = dividing a visual scene into meaningful regions or “chunks.”

Segmentation can use differences in:

• texture
• colour
• motion
• depth
• size
• shape

Grouping vs Segmentation

Grouping

“Which visual elements belong together?”

Segmentation

“Where are the separate regions/objects in this scene?”

Both –> shape representation.

Contour Integration

After detecting local edges, the visual system also needs to determine which edges connect to form an object’s outline.

Contour integration = linking local edge information together to create continuous shapes.

This allows us to perceive an object’s outline even when parts are:

• hidden
• missing
• interrupted

Example

If part of an object is hidden:

visible edge → hidden section → visible edge

the visual system can still perceive those edges as belonging to one continuous object.

Possible Mechanisms of Contour Integration

Collector units

Combine information from multiple local edge signals.

Cooperative interactions

Nearby neurons responding to compatible edges enhance each other’s activity.

Feedback from higher brain areas

Higher-level areas send information back to help refine the perceptual interpretation.

Together, these processes help construct coherent shapes from fragmented visual input.

How the Brain Builds a Shape

Step 1 — Detect Local Features

Detect:

• edges
• lines
• colours

↓

Step 2 — Gestalt Grouping

Determine which features probably belong together.

↓

Step 3 — Segment the Scene

Divide the image into meaningful regions using cues such as:

• colour
• texture
• motion

↓

Step 4 — Integrate Contours

Connect local edges into continuous boundaries.

↓

Step 5 — Unified Shape Representation

Combine the information into a coherent shape.

That shape is now ready to move into Stage 3: Object Representation, where it can be matched with stored object knowledge.