Human-computer interaction is the study of how humans and computers interact with each other and how computers can be designed to work well with humans. Over time, this field has grown to cover almost all kinds of information technology design, not just computers in the traditional sense. At its core, HCI looks at how people use computers in practice and how effectively computers are designed to support that interaction.
As it deals with both humans and computers, HCI naturally draws from several disciplines: computer science, cognitive science, psychology, and sociology.
A usable system is easy and simple to use. You don't need a manual to use it, and you don't get confused while using it. A useful system actually does something you need. It solves your problem. A system can nail one without the other. A system can be usable but not very useful.
For example, a simple calculator app that performs basic arithmetic operations with a great user interface, easy for anyone to use, but it is essentially useless to a college student who needs to solve complex calculus problems involving logarithmic functions.
In most email apps, for example Gmail, people usually just want to reply to one person but the problem is that the "Reply" and "Reply All" buttons are right next to each other and are almost identical, so it's very easy to click the wrong one without noticing. This often leads to private replies accidentally going out to an entire group, sometimes causing embarrassment at work. To fix this, I would suggest showing a quick warning when "Reply All" is clicked on a large group for example, "Send to 15 people?" giving users a chance to notice the mistake before it's sent..
A prototype is an interactive version of a design that simulates how a website or app actually works; users can actually click through, navigate, and test it, rather than just look at it. Testing a prototype with users lets a designer catch problems like confusing navigation before any real development begins. Asking "Do you like it?" will just get you an opinion; it is a yes or no question based on how the design looks or feels. It won't tell you if someone could actually complete a task, where they got stuck, or what specifically confused them.
Voice interfaces like Google Assistant or Amazon Alexa let users just speak commands and get spoken responses back. Instead of having to do typing or clicking, this is especially useful for people who are blind or visually impaired, since they can't easily rely on a screen. Assistive technologies help blind and visually impaired people manage daily tasks independently through voice, without needing to read a screen at all. You could test it with real users from this group of people, giving them specific tasks to complete by voice and measuring things like the task success rate, how long it took, and where the users got confused or needed help.