AURA — Deep Research Agent

Norman transitioned from the rigorous, often esoteric world of academic cognitive science—having founded the Cognitive Science Department at the University of California, San Diego (UCSD)—into the realm of applied design ^{xZINMdYXi0x2bq5DLRwGpppk2RdiHS4IX8Qk1NOSpkIQGJXYPntP97HPDenghAmOKZy98j9CdGcYDin-QBb9ieEXX0Z3hO2COZLSTArina4KyrIR0lNTlBuEekGStXIogzaQ==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">medium.com">11}^3]. He observed that university-based research, while profound, often had little impact on society at large ^{qj2T2GuCRVd0sFbTabVq4Gl3id-ThZXrDm7gw3SoJqpxFQ3KgBUFbqybw==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">supersummary.com">14}^1]. This realization birthed a lifelong crusade against the "psychopathology of everyday things," a term he coined to describe the frustration users experience when confronted with poorly designed objects, from unfathomable shower controls to convoluted software ^{8t-Hh7Coa4CaMoTzPxSHIUNWGU2reZMXuoidpUpXsnX1GmatipC0HEBQu0elPi8gq2Bhy4UqirMbDLaPejN3Ky2pe1nulnMrthNO6iOW-QpqnK29ViyI2gf}FIHYhUEaDuJbyFoK8eOj-ARynG9ouHBzFdKOlNW3z8PdA769uyEUiyYoYTew==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">medium.com">9^4].

Through seminal works such as The Design of Everyday Things (DOET), Emotional Design, and Design for a Better World, Norman has consistently anticipated how humans interact with the designed world. His philosophy dictates that technology must adapt to the user, not the other way around ^{7oRnduDD7rZJKm6PfvTBwBViE0yAbMmb}DbCJkpHTk6pTr5W6f1C442H2pThNhvg1u1rt-bXBiv0oEll1bLjkEA1HezxyrUtyE0XxbDonBZp8Z2C3EMjRQon4Dohb" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">frankspillers.com">10^5]. Today, as we navigate the complexities of digital banking platforms, agentic AI, and global sociotechnical systems, Norman’s principles provide an indispensable blueprint for creating usable, humane, and effective futures.

[2] The Cognitive Foundations of Design [source]

To understand Norman’s approach to design, one must first understand his roots in cognitive psychology. In the 1960s and 1970s, Norman focused his research on human perception, memory, and attention ^{mphlBCJFDLsiSP5mTnXhvMdNTalac0RqViee1b57QN60wG86m-FTA4jnYV77qOcCbvuE65GHZ6j2m72DVYHRNDRdP7n4pa7tzygU-RYkhANpJRjWPbm7AGlNDEpMyoUfeQdlAiKYcz3MA7ktHRBoNDq1MSa2Pla0fNYIbuB6L6Akbw35qpv38hA==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">medium.com">23}^6]. His work sought to build a biological and realistic account of cognitive activity, recognizing that the human brain relies heavily on context and experience to interpret the world ^{npZ1LfvpwkrHSQ-CMojG3L3SZjdCb8TOHxQMnK8rhdstFRARAoDFzLTk6BPs2Nq24J84AYOvoriCl613iprLAg==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">logicalimagination.com">12}^2].

[2] 1 The Limits of Human Memory [source]

Engineers often design systems based on pure logic, expecting users to memorize complex sequences of operations. Norman’s cognitive research demonstrated that human working memory is severely limited and highly fallible ^{AQP1Yt0ZkKh9SomamYQjNxGevyHndrA3X3AmPX4wB718PXNjDifp6eTZQeNg6lUL5o0N7U8JY8A89bXjiygpo8U=" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">bornfight.com">8}^7]. To compensate, humans rely on two types of knowledge:

Knowledge in the Head: Internalized information, such as passwords or learned procedures.
Knowledge in the World: External cues provided by the environment, such as labels, shapes, and spatial layouts ^{8t-Hh7Coa4CaMoTzPxSHIUNWGU2reZMXuoidpUpXsnX1GmatipC0HEBQu0elPi8gq2Bhy4UqirMbDLaPejN3Ky2pe1nulnMrthNO6iOW-QpqnK29ViyI2gfFIHYhUEaDuJbyFoK8eOj-ARynG9ouHBzFdKOlNW3z8PdA769uyEUiyYoYTew==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">medium.com">9}^4].

Norman argued that good design minimizes the cognitive load placed on "knowledge in the head" by embedding "knowledge in the world" directly into the interface ^{8t-Hh7Coa4CaMoTzPxSHIUNWGU2reZMXuoidpUpXsnX1GmatipC0HEBQu0elPi8gq2Bhy4UqirMbDLaPejN3Ky2pe1nulnMrthNO6iOW-QpqnK29ViyI2gfFIHYhUEaDuJbyFoK8eOj-ARynG9ouHBzFdKOlNW3z8PdA769uyEUiyYoYTew==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">medium.com">9}^4]. When a user approaches a physical or digital product, they should not need to rely on perfect recall; the design itself should guide their actions.

[2] 2 The Seven Stages of Action [source]

Norman formalized human interaction into a conceptual model known as the Seven Stages of Action. This framework explains how people attempt to bridge the gulfs between their internal goals and the physical execution of tasks:

Goal: Forming the intent.
Plan: Determining the action sequence.
Specify: Mapping out the exact steps.
Perform: Executing the physical action.
Perceive: Observing the state of the system after the action.
Interpret: Making sense of the system's new state.
Compare: Evaluating whether the goal was achieved ^{AQP1Yt0ZkKh9SomamYQjNxGevyHndrA3X3AmPX4wB718PXNjDifp6eTZQeNg6lUL5o0N7U8JY8A89bXjiygpo8U=" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">bornfight.com">8}^7].

Designers must bridge the Gulf of Execution (how do I work this?) and the Gulf of Evaluation (what just happened?) ^{AQP1Yt0ZkKh9SomamYQjNxGevyHndrA3X3AmPX4wB718PXNjDifp6eTZQeNg6lUL5o0N7U8JY8A89bXjiygpo8U=" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">bornfight.com">8}^7]. When these gulfs are wide, users experience confusion and frustration; when they are narrow, the interaction feels seamless and intuitive.

[3] Core Concepts: The Vocabulary of Interaction [source]

Norman provided the design industry with its foundational vocabulary. Before The Design of Everyday Things (originally published in 1988 as The Psychology of Everyday Things), terms like usability were abstract. Norman codified the mechanics of intuitive interaction through a set of core principles ^{uThZJcDT4h8WBSaRBeVw7Cnvtq2FE31Opy1MjGbr706BncF2SRXjR3g-lLM8xQqGXFcpoj3fCO9j3gVzENAlQylnpQ7kQrBZygfRyfp-WmliVr4xQRaPEjhVCXi56AzHlw0rbkXKB3YpZijwji5dFp0Q34rQcJ2R0YIZvhDGdwaPUSYhIfVAvBtZROb0=" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">goodreads.com">22}^8].

[3] 1 Affordances vs. Signifiers [source]

The most famous, and perhaps most misunderstood, concept in Norman's lexicon is the affordance. Borrowed from ecological psychologist James J. Gibson, an affordance is the relationship between the properties of an object and the capabilities of an agent that determine just how the object could possibly be used ^{s44J5mOu6AWm7BpllruQCHJITh7g37As8qSVKDnZl8sG9Ifj8JH7mvxnEMu8ejTixNY1CZjUR0j6TSy-mey9faMEE6Zk5we2fOfEvw5Y52tBQtNrvVyRiUCto=" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">wikipedia.org">3}^9]. For example, a chair affords sitting; a button affords pressing.

However, Norman noticed that in the digital realm, the concept was being misused. Screens have no inherent physical affordances—a digital button is merely a collection of pixels ^{s44J5mOu6AWm7BpllruQCHJITh7g37As8qSVKDnZl8sG9Ifj8JH7mvxnEMu8ejTixNY1CZjUR0j6TSy-mey9faMEE6Zk5we2fOfEvw5Y52tBQtNrvVyRiUCto=" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">wikipedia.org">3}^9]. Therefore, he introduced the concept of signifiers in the revised edition of his book.

Affordance: What actions are physically possible.
Signifier: The perceptible signal (visual, auditory, or tactile) that communicates where and how to act ^{Py7DkD9wH8L2D0kqeEDYnEEpFdu2rYQ1BdCu2R-hjsx7mxo5D1LVP" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">stanford.edu">1}^10].

As Norman states, "signifiers are more important than affordances" because they communicate usability ^{p8fLW-Thr7LhqlVKDnvEsWnmJkJvzR3UwsZbBlhteNmzGB9RjNoIBE7qcraLrC7ELE0YgC7IxS" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">rsdsymposium.org">30}^11]. A hidden door affords passage, but without a handle or a push plate (the signifier), the user will never discover it.

[3] 2 The "Norman Door" Phenomenon [source]

The quintessential demonstration of failed discoverability is the Norman Door ^{s44J5mOu6AWm7BpllruQCHJITh7g37As8qSVKDnZl8sG9Ifj8JH7mvxnEMu8ejTixNY1CZjUR0j6TSy-mey9faMEE6Zk5we2fOfEvw5Y52tBQtNrvVyRiUCto=" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">wikipedia.org">3}^9]. Walk into almost any commercial building, and you will find a door with a vertical pull handle that actually requires you to push, or a flat push plate that requires you to pull. When users invariably perform the wrong action, they feel stupid ^{bKrXjHmoEQtn9xfjGqCgrb0GtNWpBHihXcDVVuhZZ9DSQfG9YDzcD1bctVQ69KZkj918cR6CCQjGD7LLqB1Z3uTwyMWrKF57EGVcjBsaoX4VW0C9hK8jfw3DdHDB0qBA==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">ryandelaney.co">5}^12].

Norman famously pointed out that a door requiring a "Push" or "Pull" label is a symptom of fundamentally flawed design. The physical design of the door should inherently signal the correct action ^{bOydcdp8u3DOFnUvfElOJRvWQfvISilkGzJ0KWstf5gNqxm4jDHobW6f8f4Wy0K6C9mOGYcJTJx3U9EQDIb8zW7ruy2ZzlDeXrYIPZVs}ovOWNEG9s8=" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">youtube.com">2^13]. The Norman Door serves as a powerful metaphor for software interfaces where users are blamed for clicking the wrong button when the interface itself provided conflicting signifiers.

[3] 3 Constraints, Mapping, and Feedback [source]

To create intuitive experiences, Norman outlines several other critical mechanisms:

Constraints: Design elements that gently force the desired behavior by making wrong actions impossible or difficult (e.g., a USB plug that only fits one way, or a greyed-out digital button) ^{bOydcdp8u3DOFnUvfElOJRvWQfvISilkGzJ0KWstf5gNqxm4jDHobW6f8f4Wy0K6C9mOGYcJTJx3U9EQDIb8zW7ruy2ZzlDeXrYIPZVsovOWNEG9s8=" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">youtube.com">2}^13].
Mapping: The spatial and logical relationship between controls and their effects. A classic example is the Mercedes-Benz seat control, shaped exactly like the seat it adjusts. Good mapping reduces the cognitive translation required to operate a system ^{uA6P7MFdmOzOs1ZpHiQBvlFZzfmtTC-wAaKBP6kzqycxcH0Mo839GPwQNkBWzonnES40rBevfbWf8mJy8A2kILcculHCZr4gYtghybcFPPK4EDlNakc0AT4KndzQ==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">getabstract.com">18}^14].
Feedback: Full and continuous information about the results of actions. Without immediate feedback (a loading spinner, a tactile click, a confirmation message), users are left in a state of anxiety, unsure if their action registered ^{bKrXjHmoEQtn9xfjGqCgrb0GtNWpBHihXcDVVuhZZ9D}SQfG9YDzcD1bctVQ69KZkj918cR6CCQjGD7LLqB1Z3uTwyMWrKF57EGVcjBsaoX4VW0C9hK8jfw3DdHDB0qBA==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">ryandelaney.co">5^12].

[3] 4 Conceptual Models [source]

A strong conceptual model is the overarching mental framework a user builds about how a system works. Designers create a "system image" through the product's interface, manuals, and physical form ^{eOp0wwWVqFg14Dnw6B2l-O-9d4RWZuFggmvV3i--ojDfQckP-k2BslUnoRbdO5HCeNYcb8VYhuzbfgqp9bbT6dUbCPKB7OqH3rl44XphOAEqTJM8ZV3ZXKRq}0KpxwopukcmKkvUAfMgZUirF1TDC6QA4PGzcbi1OiQzmcy5XJFSnJut" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">wordpress.com">24^15]. If the system image is clear, the user develops an accurate conceptual model and can intuitively navigate new features without instruction ^{7oRnduDD7rZJKm6PfvTBwBViE0yAbMmbDbCJkpHTk6pTr5W6f1C442H2pThNhvg1u1rt-bXBiv0oEll1bLjkEA1HezxyrUtyE0XxbDonBZp8Z2C3EMjRQon4Dohb" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">frankspillers.com">10}^5].

[4] The Psychology of Human Error [source]

One of Norman's most transformative contributions is his destigmatization of human error. In high-stakes environments—aviation, nuclear power, medical devices—accidents are routinely blamed on "operator error."

[4] 1 The Myth of the Careless User [source]

Norman highlights that an estimated 75% to 95% of industrial accidents are blamed on "human error" ⁵⁶^16]. However, Norman argues that high failure rates are a symptom of badly designed systems that require unnatural precision, flawless memory, or perfect attention over long periods ^{AQP1Yt0ZkKh9SomamYQjNxGevyHndrA3X3AmPX4wB718PXNjDifp6eTZQeNg6lUL5o0N7U8JY8A89bXjiygpo8U=" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">bornfight.com">8}^7]. If a system permits or induces a user to make an error, the fault lies with the design, not the user ^{eOp0wwWVqFg14Dnw6B2l-O-9d4RWZuFggmvV3i--ojDfQckP-k2BslUnoRbdO5HCeNYcb8VYhuzbfgqp9bbT6dUbCPKB7OqH3rl44XphOAEqTJM8ZV3ZXKRq_0KpxwopukcmKkvUAfMgZUirF1TDC6QA4PGzcbi1OiQzmcy5XJFSnJut" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">wordpress.com">24}^15].

He advocates for rigorous root-cause analysis—asking "why" repeatedly until the underlying design flaw is uncovered ^{8t-Hh7Coa4CaMoTzPxSHIUNWGU2reZMXuoidpUpXsnX1GmatipC0HEBQu0elPi8gq2Bhy4UqirMbDLaPejN3Ky2pe1nulnMrthNO6iOW-QpqnK29ViyI2gf}FIHYhUEaDuJbyFoK8eOj-ARynG9ouHBzFdKOlNW3z8PdA769uyEUiyYoYTew==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">medium.com">9^4]. Machines do not have common sense; they require precise inputs. Humans, conversely, are flexible, imaginative, and prone to distraction. Designing systems that force humans to act like machines is a recipe for disaster ^{1st6DODV034FiJfBeWVq1TUL29skF04l1qyIYM4cGfhmReGQX2TkAUZ-dtvva2Ol9pXMGeF6QXdbdrLJFK2KQiBPYdL5zaM72RBKAR4MPK5hMVFunWIM9dRVHrzcka0yJpnIr3aJ7CgFJ0xu6NuJxo5Vqf" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">muz.li">27}^17].

[4] 2 Slips vs. Mistakes [source]

Working with psychologist James Reason, Norman classified errors into two primary categories, which inform how interfaces should be designed to mitigate them:

Slips: The user forms the correct goal but executes the wrong action.

Action-based slips: Hitting the wrong key on a keyboard.
Memory-lapse slips: Forgetting to attach a file before hitting "send."
Design solution: Implement constraints, confirmation dialogues, and clear feedback.

Mistakes: The user forms the wrong goal or plan, often due to a flawed conceptual model.

Rule-based mistakes: Applying the wrong procedure to a situation.
Knowledge-based mistakes: Misunderstanding the state of the system.
Design solution: Improve system visibility and mapping to help users build accurate mental models ⁵⁶^16].

[5] Emotional Design: Beyond Utility [source]

In 2004, Norman published Emotional Design: Why We Love (or Hate) Everyday Things. This marked a significant evolution in his thinking. He acknowledged that his earlier work focused almost exclusively on logic, utility, and function, sometimes leading to the critique that DOET promoted highly usable but aesthetically sterile products ^{uA6P7MFdmOzOs1ZpHiQBvlFZzfmtTC-wAaKB}P6kzqycxcH0Mo839GPwQNkBWzonnES40rBevfbWf8mJy8A2kILcculHCZr4gYtghybcFPPK4EDlNakc0AT4KndzQ==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">getabstract.com">18^14].

Norman realized that human cognition and emotion are inextricably linked. "Cognition attempts to make sense of the world: emotion assigns value" ⁵⁹^7].

[5] 1 The Aesthetic-Usability Effect [source]

Norman popularized the aesthetic-usability effect, positing that "attractive things work better" ^{fnuYJTwVgYsLQaLqo8fkiytC7hV1jYG7jtXd9TkGPM7tANkAPU9vK64ufx2w}ageEmTNDMbWMpxDdcpa9OPgghHqtekm7cxSw8QtdazUnPZvxOit-KbrNHYNsjQIfT71JVreOiuodJbuVGw4pHg9GENRB3eYl1cA==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">shortform.com">19^18]. Positive emotional states—induced by pleasing aesthetics—relax the brain, making users more creative and more tolerant of minor design flaws ^{j1mNBzjroCPBIkn0tOaPUKJinN9GaN1pJ-RdwMxzgpmVBJiMQWc7gJ62w1aen2m5-j0iHCiB--VBWryS62iEAYkV24kBoHeaESDkU96vSFySflI4DmfuVTE8OS6wJVSJ3X7FLqhcRkvalMcskwFdy-f539j5Nd8GhWiDrg==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">howtoes.blog">16}^19]. Conversely, negative emotions trigger tunnel vision, causing users to fixate on problems and struggle with tasks they might otherwise easily complete ^{j1mNBzjroCPBIkn0tOaPUKJinN9GaN1pJ-RdwMxzgpmVBJiMQWc7gJ62w1aen2m5-j0iHCiB--VBWryS62iEAYkV24kBoHeaESDkU96vSFySflI4DmfuVTE8OS6wJVSJ3X7FLqhcRkvalMcskwFdy-f539j5Nd8GhWiDrg==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">howtoes.blog">16}^19].

[5] 2 The Three Levels of Emotional Processing [source]

Norman divided emotional design into three distinct cognitive levels, each requiring a different design strategy ^{TmrhHNQfu6cxkuUsC7YZY9DXt7LG88ZKIbSuLXjlYp0THX45eAZkA7fubA73TTAZ33mHmKZEEqGZ39aF9jKe61iE2YG95O}SAhcvekr-fqOC1afsaRfdDYoxp4adtsLzVpdQnqJnhItt7Me11Yg9Ry1srM6fzQiPvg==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">muz.li">20^20]:

Level	Cognitive Processing	Focus of Design	Real-World Example
Visceral	Pre-conscious, automatic, biological. Driven by immediate sensory input.	Appearance, shape, color, initial tactile feel. "Love at first sight."	The sleek curves and premium heft of an iPhone; the satisfying click of a mechanical keyboard ⁷⁴^21].
Behavioral	Subconscious, routine. Driven by the physical execution of tasks.	Usability, performance, function, feedback.	The intuitive mapping of a car's steering wheel; the seamless one-click checkout on Amazon ⁷¹^22].
Reflective	Conscious, rational, intellectual. Driven by self-image, memories, and culture.	Meaning, brand identity, prestige, emotional connection over time.	Owning a vintage mechanical watch; the pride associated with driving an eco-friendly Tesla ^{uA6P7MFdmOzOs1ZpHiQBvlFZzfmtTC-wAaKBP6kzqycxcH0Mo839GPwQNkBWzonnES40rBevfbWf8mJy8A2kILcculHCZr4gYtghybcFPPK4EDlNakc0AT4KndzQ==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">getabstract.com">18}^14].

To achieve true user delight, a product must succeed across all three levels. A product might be viscerally striking but behaviorally frustrating (a beautiful but slow website), or behaviorally excellent but reflectively dull (a utilitarian spreadsheet tool) ^{TmrhHNQfu6cxkuUsC7YZY9DXt7LG88ZKIbSuLXjlYp0THX45eAZkA7fubA73TTAZ33mHmKZEEqGZ39aF9jKe61iE2YG95OSAhcvekr-fqOC1afsaRfdDYoxp4adtsLzVpdQnqJnhItt7Me11Yg9Ry1srM6fzQiPvg==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">muz.li">20}^20].

[6] Real-World Applications: Financial Services & Complex Systems [source]

For Design Leaders in complex industries like Financial Services, Norman’s principles are not theoretical luxuries; they are operational imperatives. The banking sector handles the most emotionally charged aspect of a user's life: their financial security ⁴⁰^23].

[6] 1 Cognitive Load and Financial Anxiety [source]

In banking UX, cognitive load directly correlates with user anxiety. When a user logs into a digital banking app to transfer funds urgently, a cluttered interface lacking clear signifiers acts as a psychological barrier ⁴⁰^23]. As Norman notes, the whole point of human-centered design is to tame complexity ⁷²^24].

Capital One serves as a premier case study in behavioral and emotional design within finance. Their mobile app prioritizes the reduction of cognitive load by surfacing only essential items—balances, recent transactions, and payments—with a clean, uncluttered design. By avoiding "feature bloat" on the home screen, they create a sense of control and calmness ⁵⁵^23]. Furthermore, features like CreditWise are integrated seamlessly, utilizing clear feedback to build user trust and reinforce positive financial habits ⁵⁵^23].

[6] 2 Designing for Diverse Populations [source]

Norman's concept of cultural constraints is vital when designing global financial products. A recent case study analyzing mobile banking applications in South Korea demonstrated the profound impact of poor usability on international users. The study found that international students took an average of 35% longer to complete standard banking tasks compared to local users due to navigation complexities and lack of culturally adaptable UI/UX elements ⁴⁵^25].

When interfaces fail to align with the user's cultural mental models, error rates skyrocket. If a banking app's load time exceeds 2 seconds, the perceived delay feels 35% longer to the user, leading to a 48% uninstallation rate ⁶⁴^26]. In finance, this friction does not just cause annoyance; it causes transaction abandonment and erodes institutional trust.

[6] 3 Emotional Design in Banking [source]

The UXDA (User Experience Design Agency) applies Norman's emotional design principles specifically to banking. They note that the traditional banking mindset assumes users are rational actors seeking maximum features ⁴³^27]. In reality, users suffer from "decision paralysis" when presented with too many options. By applying visceral design (clean, modern aesthetics) and behavioral design (intuitive money transfers mimicking physical cash handoffs), challenger banks like Monzo and Ally have disrupted legacy institutions. Monzo's instant notifications provide immediate systemic feedback, transforming the historically anxiety-inducing wait for a transaction to clear into a reassuring, transparent experience ⁴⁰^23].

[7] The Evolution: From Human-Centered to Humanity-Centered Design [source]

In 2023, coming out of retirement for the fifth time, Norman published Design for a Better World: Meaningful, Sustainable, Humanity Centered ^{Fim11ok8aRs461dwjLlZWLwHkTOOUJffTuYreBQDeKBZw8nN3KkeexEhLGISfoJr4iayC6eIOeUDuQr1vjx8AQWpm2oGQuPYOEANUClehshJhxdCQlMJ07s8Ip4GT19ltKHs=" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">aimind.so">32}^28]. This work represents a radical expansion of his philosophy, shifting the focus from the individual "user" to the broader ecosystem of "humanity."

[7] 1 Critiques of Traditional HCD [source]

Norman acknowledged that traditional Human-Centered Design (HCD), which he helped popularize, has unintended negative consequences. By hyper-focusing on optimizing experiences for individual consumers, HCD often ignores the long-term systemic impacts on society and the environment ^{SYdjB613QzU7ac5hL5wrVBnhl0pJ2s5QwfJOTttgwCV0h53qAldlTm-DFegdnpFBicT1gwQaqB6yM4=" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">stabilarity.com">33}^29]. It drives consumerism, waste, and ecological degradation ^{wMMs12U5qktxrb-fxDas2Gh6usTPcuktvo7OKgdCyxx2sgpGinye7QlfpA39AUiRpLBjlEcMuE1AudsB0DN5gPPttYKMh-26CorxiLeICLKLCvxmS8Peo4Ck-UtqK9l86DR1WA==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">substack.com">34}^30].

[7] 2 The Five Tenets of Humanity-Centered Design [source]

Norman advocates for aligning design efforts with the United Nations Sustainable Development Goals (SDGs) and proposes five core tenets for Humanity-Centered Design:

Address the Root Problem: Look beyond the immediate brief to solve underlying systemic issues.
Focus on the Ecosystem: Consider the impact on all stakeholders, including the environment and non-user populations.
Take a Long-Term Systems Perspective: Recognize that interventions in complex networks have downstream effects.
Continually Test and Refine: Embrace iterative, muddling-through progress rather than seeking perfect, static solutions.
Design With the Community: Move away from designing for people to designing with people, empowering local communities to solve their own problems ^{SYdjB613QzU7ac5hL5wrVBnhl0pJ2s5QwfJOTttgwCV0h53qAldlTm-DFegdnpFBicT1gwQaqB6yM4=" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">stabilarity.com">33}^29].

This shift demands that design leaders—especially those in capital-intensive sectors like finance—recalibrate their metrics of success from mere engagement and profit to sustainability and societal well-being ^{wMMs12U5qktxrb-fxDas2Gh6usTPcuktvo7OKgdCyxx2sgpGinye7QlfpA39AUiRpLBjlEcMuE1AudsB0DN5gPPttYKMh-26CorxiLeICLKLCvxmS8Peo4Ck-UtqK9l86DR1WA==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">substack.com">34}^30].

[8] Agentic UX, AI, and the Future of Interaction [source]

As artificial intelligence rapidly permeates digital products, Norman's principles are being severely tested. The rise of Generative AI and "Agentic AI" (systems capable of autonomous decision-making with minimal oversight) has created an unprecedented UX crisis.

[8] 1 The "Black Box" Problem [source]

Norman has repeatedly warned against "black box" technologies—systems that make decisions without providing explanations to the user ^{1rLxLtwj1IkaAMVDryPDFNbLrkTFttX42RMbGHz97mXnBWvaTLeeSv-xOzbREfQL5aUX56ZoIMXsfzVpq3qboukrVf6AE9nCcW6jTQ==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">miquido.com">26}^31]. Contemporary AI often functions exactly as a black box. When an AI generates a response or executes a trade, the user has no visibility into its internal state, its memory, or its logic ³⁸^32].

This opacity directly violates Norman's core principles of visibility, feedback, and conceptual models. If users cannot understand how an AI reaches a conclusion, they cannot build an accurate mental model, and consequently, trust is entirely eroded ^{1rLxLtwj1IkaAMVDryPDFNbLrkTFttX42RMbGHz97mXnBWvaTLeeSv-xOzbREfQL5aUX56ZoIMXsfzVpq3qboukrVf6AE9nCcW6jTQ==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">miquido.com">26}^31].

[8] 2 Enterprise AI Failure Rates [source]

The consequences of ignoring foundational UX principles in AI development are financially catastrophic. Recent data reveals that Enterprise AI initiatives fail at rates between 80% and 95% ⁶⁵^33].

A RAND Corporation study confirmed AI project failure rates exceed 80%, double the rate of non-AI IT projects ⁶⁷^34].
An MIT NANDA study (2025) reported that 95% of enterprise AI pilots fail to reach production ⁶⁸^35].

While much of the industry blames technical algorithms or data quality, UX researchers point to "pre-Norman" design failures ³⁸^32]. For instance, Air Canada faced legal repercussions when its customer service chatbot hallucinatively promised bereavement fares; the system provided "feedback" (an assurance to the user) but its internal state was entirely disconnected from actual policy ³⁸^32].

[8] 3 Applying Norman's Playbook to Agentic UX [source]

To rescue AI from these failure rates, designers must integrate Norman's principles into "Agentic UX" ³⁸^32]:

Norman Principle Failure in Current AI Agentic UX Solution
Visibility AI context and memory reset or are hidden from the user. Dispositional Modeling: Make the AI's "memory" and parameters visible so users know what context is being applied ³⁸^32].
Feedback AI provides confident answers without showing its work (hallucinations). Contextual Continuity: UI features like Perplexity’s citation links or ChatGPT’s "Search" steps that expose the reasoning process ⁷⁶^36].
Conceptual Model Generative models are unpredictable; users cannot predict outputs. Explainable AI (XAI): UI overlays that explain why an agentic action was recommended, allowing humans to safely supervise ⁷⁸^37].
Collaboration AI is treated as an autonomous replacement for human effort. Human-in-the-Loop: Treat interaction as a collaboration where AI augments human capability and allows user overrides ^{1rLxLtwj1IkaAMVDryPDFNbLrkTFttX42RMbGHz97mXnBWvaTLeeSv-xOzbREfQL5aUX56ZoIMXsfzVpq3qboukrVf6AE9nCcW6jTQ==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">miquido.com">26}^31].

Norman emphasizes that the goal of technology is to "tame complexity" and amplify human capability ⁷²^24]. The "Daft Punk problem" in current AI development—focusing purely on making models harder, better, faster, stronger—misses the human element. Session-based AI that forces humans to repeatedly re-explain context makes the human work harder, fundamentally reversing the purpose of human-centered design ³⁸^32].

[9] Norman's Lasting Influence and Critiques [source]

Don Norman’s legacy is not without its challenges and critiques. His early advocacy for rigorous user research led some organizations into "analysis paralysis," while his later focus on humanity-centered design has been criticized by some pragmatists as overly idealistic in the face of entrenched corporate capitalism ⁵⁰^28].

Furthermore, Norman is deeply critical of modern design's tendency toward "first-version hubris"—the belief that a product can be perfectly designed in a vacuum prior to release ^{1rLxLtwj1IkaAMVDryPDFNbLrkTFttX42RMbGHz97mXnBWvaTLeeSv-xOzbREfQL5aUX56ZoIMXsfzVpq3qboukrVf6AE9nCcW6jTQ==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">miquido.com">26}^31]. He argues strongly for the iterative nature of design, championing the "fail often, fail fast" mentality to learn from users in real-world environments ^{1st6DODV034FiJfBeWVq1TUL29skF04l1qyIYM4cGfhmReGQX2TkAUZ-dtvva2Ol9pXMGeF6QXdbdrLJF}K2KQiBPYdL5zaM72RBKAR4MPK5hMVFunWIM9dRVHrzcka0yJpnIr3aJ7CgFJ0xu6NuJxo5Vqf" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">muz.li">27^17]. He coined "Norman's Law of Product Development," which states: "The day a product development process starts, it is behind schedule and above budget" ^{1st6DODV034FiJfBeWVq1TUL29skF04l1qyIYM4cGfhmReGQX2TkAUZ-dtvva2Ol9pXMGeF6QXdbdrLJFK2KQiBPYdL5zaM72RBKAR4MPK5hMVFunWIM9dRVHrzcka0yJpnIr3aJ7CgFJ0xu6NuJxo5Vqf" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">muz.li">27}^17].

Despite these critiques, his frameworks have proven incredibly resilient. Whether applied to the physical design of hospital signage, the digital layout of an online trading platform, or the conversational interface of a Large Language Model, the triad of discoverability, understanding, and emotional resonance remains the gold standard for product success.

[10] Conclusion: The Blueprint for Tomorrow's Interfaces [source]

Don Norman did not just write the rulebook for usability; he provided the philosophical foundation for how humans and machines ought to coexist. By shifting the blame for operational failures from the "erratic human" to the "poorly designed system," he forced technologists to develop empathy for the end user ^{8t-Hh7Coa4CaMoTzPxSHIUNWGU2reZMXuoidpUpXsnX1GmatipC0HEBQu0elPi8gq2Bhy4UqirMbDLaPejN3Ky2pe1nulnMrthNO6iOW-QpqnK29ViyI2gfFIHYhUEaDuJbyFoK8eOj-ARynG9ouHBzFdKOlNW3z8PdA769uyEUiyYoYTew==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">medium.com">9}^4].

As we move deeper into an era characterized by complex sociotechnical systems, autonomous financial algorithms, and omnipresent AI, the temptation to prioritize technological capability over human usability is strong. Yet, as the 80-95% failure rate of enterprise AI clearly illustrates, raw computational power is useless if humans cannot discover its functions, understand its actions, or trust its outcomes ⁶⁵^33].

For design leaders, product managers, and engineers, Norman's work serves as a vital reminder: technology changes rapidly, but human psychology changes at a glacial pace ^{7oRnduDD7rZJKm6PfvTBwBViE0yAbMmbDbCJkpHTk6pTr5W6f1C442H2pThNhvg1u1rt-bXBiv0oEll1bLjkEA1HezxyrUtyE0XxbDonBZp8Z2C3EMjRQon4Dohb" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">frankspillers.com">10}^5]. By adhering to his principles of affordances, signifiers, systemic feedback, and emotional design, we can continue to design futures that are not only highly functional but profoundly humane.

References

Sources:

stanford.edu
youtube.com
wikipedia.org
melissadu.com
ryandelaney.co
smallyu.net
readingraphics.com
bornfight.com
medium.com
frankspillers.com
medium.com
logicalimagination.com
tianpan.co
supersummary.com
sahandmalaei.com
howtoes.blog
wordpress.com
getabstract.com
shortform.com
muz.li
softway.net
goodreads.com
medium.com
wordpress.com
davidpublisher.com
miquido.com
muz.li
designresearchsociety.org
arxiv.org
rsdsymposium.org
newartschool.education
aimind.so
stabilarity.com
substack.com
medium.com
medium.com
arxiv.org

Don Norman: Designing Usable Futures

[2] The Cognitive Foundations of Design [source]

[2] 1 The Limits of Human Memory [source]

[2] 2 The Seven Stages of Action [source]

[3] Core Concepts: The Vocabulary of Interaction [source]

[3] 1 Affordances vs. Signifiers [source]

[3] 2 The "Norman Door" Phenomenon [source]

[3] 3 Constraints, Mapping, and Feedback [source]

[3] 4 Conceptual Models [source]

[4] The Psychology of Human Error [source]

[4] 1 The Myth of the Careless User [source]

[4] 2 Slips vs. Mistakes [source]

[5] Emotional Design: Beyond Utility [source]

[5] 1 The Aesthetic-Usability Effect [source]

[5] 2 The Three Levels of Emotional Processing [source]

[6] Real-World Applications: Financial Services & Complex Systems [source]

[6] 1 Cognitive Load and Financial Anxiety [source]

[6] 2 Designing for Diverse Populations [source]

[6] 3 Emotional Design in Banking [source]

[7] The Evolution: From Human-Centered to Humanity-Centered Design [source]

[7] 1 Critiques of Traditional HCD [source]

[7] 2 The Five Tenets of Humanity-Centered Design [source]

[8] Agentic UX, AI, and the Future of Interaction [source]

[8] 1 The "Black Box" Problem [source]

[8] 2 Enterprise AI Failure Rates [source]

[8] 3 Applying Norman's Playbook to Agentic UX [source]

[9] Norman's Lasting Influence and Critiques [source]

[10] Conclusion: The Blueprint for Tomorrow's Interfaces [source]

References

Norman Principle	Failure in Current AI	Agentic UX Solution
Visibility	AI context and memory reset or are hidden from the user.	Dispositional Modeling: Make the AI's "memory" and parameters visible so users know what context is being applied ³⁸^32].
Feedback	AI provides confident answers without showing its work (hallucinations).	Contextual Continuity: UI features like Perplexity’s citation links or ChatGPT’s "Search" steps that expose the reasoning process ⁷⁶^36].
Conceptual Model	Generative models are unpredictable; users cannot predict outputs.	Explainable AI (XAI): UI overlays that explain why an agentic action was recommended, allowing humans to safely supervise ⁷⁸^37].
Collaboration	AI is treated as an autonomous replacement for human effort.	Human-in-the-Loop: Treat interaction as a collaboration where AI augments human capability and allows user overrides ^{1rLxLtwj1IkaAMVDryPDFNbLrkTFttX42RMbGHz97mXnBWvaTLeeSv-xOzbREfQL5aUX56ZoIMXsfzVpq3qboukrVf6AE9nCcW6jTQ==" class="text-muted hover:text-primary border-b border-dotted border-grid-line" target="_blank" rel="noopener">miquido.com">26}^31].