Design Psychology
Key question
How much can humans process?
People have limited attention and working memory. Design can respect these limits by structuring information, using familiar patterns, and making errors forgiving. It can also overwhelm people.
Learning objective: You design information to reduce cognitive load, and evaluate interfaces using established criteria.
Good design saves thinking.
5.1 / Cognitive load
Working memory is a tight bottleneck.
Working memory only holds information briefly and in small quantities while we process it – such as a phone number on the way to the phone. It is the bottleneck of thought.
Cognitive load theory (John Sweller) distinguishes, among other things, between two types:
CONTENTS
Intrinsic
Stems from the complexity of the content itself. It can hardly be designed away, but it can be broken down into manageable chunks.
DISPLAY
Extrinsic
Caused by unclear structure, distractions, or unnecessary jargon. Design can make a real difference here.
Design can reduce extrinsic cognitive load: eliminate distractions, present steps sequentially, use familiar terms, and provide information right where it is needed.
Cut what merely distracts, not what explains.
5.2 / Chunking
Groups are easier to remember than individual parts.
Chunking means grouping individual pieces of information into meaningful units. A telephone number split into groups or an IBAN in groups of four is easier to read and remember than a long string of numbers.
George Miller's often-cited '7 ± 2' rule (1956) is a gross oversimplification. More recent research, such as that by Nelson Cowan (2001), puts the capacity of working memory closer to around four items. What matters most is how effectively information can be grouped into meaningful chunks.
In terms of design, this means: group related information together and give each group a clear heading. Avoid rigid numerical rules such as "a maximum of seven menu items", and instead check whether the groups make sense to users.
Meaningful groups beat fixed numbers.
5.3 / Hick, Fitts, Jakob
How quickly you make decisions and hit your targets can be shaped.
01
Hick's Law
The time it takes to make a decision increases with the number of options. This is not a call to cut everything down to two options, but rather to: structure options sensibly, offer default settings, and highlight what is important.
02
Fitts's Law
The time taken to hit a target depends on its distance and size. Large, nearby targets can be hit more quickly. On touchscreens, small buttons are prone to errors.
03
Jakob's Law
According to Jakob Nielsen: Users spend most of their time on other sites, and expect yours to work similarly. Familiar patterns save mental effort; any deviations should have a good reason.
Fewer options, bigger goals, familiar patterns.
5.4 / Mental Models
People operate what they believe they understand.
A mental model is an idea of how something works. It develops through experience and does not necessarily have to be correct. Design should align the users' model with the system's actual behaviour.
Don Norman distinguishes between affordances and signifiers. An affordance is the possibility of taking an action that an object offers to a person. A signifier is the perceivable clue that shows where and how to act, such as a handle or a labelled button.
The classic example is the door where, despite having a handle, you don't know whether to pull or push. Feedback is also part of this: the system shows immediately that an action has been received.
If you have to explain how it works, the design isn't finished yet.
5.5 / Heuristics
A checklist identifies problems before users do.
In 1994, Jakob Nielsen set out ten general principles for user interface design. They are called heuristics because they are broad rules of thumb rather than rigid laws:
HEURISTIC 01
Visibility of system status
The system always shows what is happening, providing prompt feedback. Example: progress bar when uploading.
HEURISTIC 02
Real-world relevance
Use the language and concepts of the users rather than technical jargon, presenting information in a natural order. For example: a recycle bin for deleting.
HEURISTIC 03
Control and Freedom
A clearly visible emergency exit: Undo, Cancel, Back. Example: "Undo" after deleting an email.
HEURISTIC 04
Consistency and standards
The same elements share the same name and appearance, adhering to familiar conventions. Example: Links are the same colour throughout.
HEURISTIC 05
Error prevention
Even better than good error messages: prevent errors from happening in the first place, or ask for confirmation before high-consequence actions. For example: use a date picker instead of free text.
HEURISTIC 06
Recognition rather than recall
Make options and information visible rather than requiring users to remember them. Example: show recently searched terms.
HEURISTIC 07
Flexibility and efficiency
Shortcuts for experienced users, without overwhelming beginners. For example: keyboard shortcuts or favourites.
HEURISTIC 08
Minimalist design
No unnecessary information: every additional element competes with what is important. Example: a form with only the essential fields.
HEURISTIC 09
Help with errors
Error messages in plain English, explaining the cause and suggesting a solution. For example: "Passwords must be at least 8 characters long" instead of "Error 403".
HEURISTIC 10
Help and documentation
Ideally intuitive to use without any explanation. If help is needed, it should be easy to find, concise, and relevant to the task. For example: a tip directly next to the input field.
The heuristics serve well as an evaluation framework. In a heuristic evaluation, several people examine an interface independently of one another, assign each identified issue to a heuristic, and sort the issues by severity.
Checking before publishing saves users the hassle.
Exercise
Apply, evaluate, transfer
Task: Heuristic evaluation of an existing website or app in small groups.
STEP 1
Choose an interface and a typical task, such as creating an account or buying a ticket.
STEP 2
Go through the task individually and note down every problem along with the heuristic violated.
STEP 3
Rate each issue according to severity (minor, medium, major) and compile all issues into a list.
STEP 4
Merge the lists and determine the three most important issues.
Evaluation: Which issues were identified independently by multiple people? Which heuristic was violated most frequently? Which improvement will deliver the greatest benefit?