In Erlang, there are two types of processes: system processes and user processes. These processes serve different purposes and have different characteristics.

1. System Processes : System processes are built-in processes that are responsible for managing various system tasks and services. They are created and managed by the Erlang runtime system. Some examples of system processes include the process scheduler, memory allocator, and I/O server. System processes typically have a lower process identifier (PID) range compared to user processes.

2. User Processes : User processes are lightweight concurrent units of execution in Erlang. They are created by the user to perform specific tasks or computations. User processes are the primary means of achieving concurrency in Erlang. Each user process has its own isolated state and execution context. User processes are created and managed by the developer using the `spawn/1` or `spawn/3` functions.

To create a user process, the `spawn/1` or `spawn/3` function is used, which takes a function as an argument. The specified function defines the behavior and logic of the process. The `spawn/1` function takes a single argument, the function that will be executed by the process, while the `spawn/3` function allows passing additional arguments to the function. Here's an example of creating a user process in Erlang :

start_process() ->
    Pid = spawn(fun() ->
        % Process behavior and logic here
        io:format("Hello from process!~n")
    end),
    Pid.​

In the above example, the `spawn/1` function is used to create a new user process. The anonymous function passed as an argument will be executed by the newly created process. The `io:format/1` function is called within the process to print a message.

The `Pid` variable captures the process identifier (PID) of the newly created process, which can be used to communicate with or monitor the process.