What is a User Interface?

Many moons ago, I attended the Georgia Institute of Technology, majoring in Computer Science. My father also attended Georgia Tech, and advised me that the Institute is "intellectual boot camp." Georgia Tech students don't say "I graduated," they say "I got out." In spite of the unrelenting pain visited upon my soul, I enjoyed my time at Georgia Tech.

As a CS major, I had the opportunity to expand my studies through specializations and electives. Human Computer Interaction was my first choice, as I have always been fascinated by user experience. In addition, I studied Cognitive Science, which was a very new program at the time. HCI and Cognitive Science are highly complementary disciplines, as Cognitive Science studies the mind as a complex amalgam of interconnected faculties, and HCI studies how computer interfaces can best align to human understanding.

HCI and UX are in the midst of a significant disruption as user expectations have shifted thanks to GenAI. In fact, I think it's time to ask a philosophical question: what is a user interface?

What is a User Interface?

Humans interact with technology all the time. But the types of interactions have dramatically shifted over the years. Computers began as devices for a very narrow portion of the population, with obtuse, complex interfaces. No longer are users expected to fiddle with punch cards, write code, or interact with textual UI on the command line to take advantage of computers.

Today, graphical interfaces have completely democratized computing, affording users simple, easy-to-understand UIs designed specifically for "WIMP" (Windows, Icons, Mouse, Keyboard) interactivity. For some categories of users, text-based interfaces still play a big role in their daily computing, and people with disabilities often rely upon assistive tools like text-to-speech, speech-to-text, audio navigation, and switch control.

The evolution from command line to GUI represented a fundamental cognitive shift. Command line interfaces required users to recall specific commands and syntax from memory. GUIs changed this paradigm entirely, allowing users to recognize options presented visually and simply point and click. This shift from recall to recognition dramatically lowered the cognitive load required to use computers.

So what is a user interface? Traditionally, it's been a mediator—a layer between human intention and machine capability. But conversational interfaces are blurring this definition in fascinating ways.

Conversational UIs: Beyond Recognition and Recall

While GUIs aren't going to disappear anytime soon, the new interaction models enabled by GenAI often don't require a graphical interface at all. Users are increasingly expecting conversational interfaces. But here's where things get interesting: conversational interfaces represent yet another cognitive shift, one that's perhaps more profound than the move from command line to GUI.

Unlike GUIs that rely on recognition, or command lines that demand precise recall, conversational interfaces require users to articulate their needs in natural language. This seems easier on the surface—after all, we converse naturally every day. But it introduces new complexities. How do users discover what's possible when there's no menu to browse? How do they recover from misunderstandings when the interface is a black box? How do we maintain user trust when the interface can reason, but we can't see how?

Consider successful conversational interfaces like voice assistants asking "What would you like to know?" versus a well-designed mobile app. The voice assistant offers infinite possibilities but zero discoverability. The app constrains possibilities but makes options explicit. This tension between capability and clarity is central to conversational UX design.

When the interface can think, reason, and even exhibit creativity, does it stop being just an interface? We're moving toward a world where the boundary between interface and intelligence becomes increasingly blurred.

The Cognitive Science Imperative

HCI researchers have studied GUIs for decades, and while core UX principles still apply to conversational interfaces, the bulk of existing research focuses elsewhere. When designing conversational interfaces, Cognitive Science becomes extremely valuable, as it touches on so many facets of the human mind—information processing, reasoning, memory, language, and emotion.

Understanding how humans process language, form mental models, and manage cognitive load isn't just helpful for conversational interface design—it's essential. As GenAI continues to permeate computing, technology vendors will need to better understand human cognition to design highly usable conversational user experiences.

The challenge ahead isn't just technical—it's deeply human. How do we design conversations that feel natural but remain purposeful? How do we handle the inherent ambiguity of human language while maintaining system reliability? These questions require insights from psychology, linguistics, and cognitive science as much as computer science.

Recommendations

Want to learn more about Cognitive Science? Here are some recommendations to get you started:

• Consciousness Explained by Daniel C. Dennett is an oldie, but a goodie. While the information within isn't the "state of the art," it is a very solid foundation to build upon.
• Mind: Introduction to Cognitive Science, 2nd Ed. is required reading. The first edition was the first book that I read on the subject of Cognitive Science.

The future of human-computer interaction will require us to understand not just how to build intelligent systems, but how to make them truly conversational. And that understanding starts with the human mind itself.