I understand that either nonelectromechanical computing devices (analog)
or electromechanical computing devices (digital) were ever developed by humans.
That distinction is a bit off, mixing unrelated categorisation. Analogue computers can be as well mechanical, as electromechanical as well as electronic (or pneumatic or hydoponic) - similar digital ...
Switches and lights.
Here's the panel for an IBM 650:
The top row of dials are used to configure in decimal values (the IBM 650 being a decimal machine). The repeated columns of 5 lights with 2 on top indicate 1 of 10 values for each digit of the 10 digit word the IBM 650 used. The lights below that row indicate operation and address. The other dials ...
From the point-of-view of someone at the time, you interacted with early computers in the usual way.
From the 1930's onwards, business and some science used IBM and Remmington-Rand mechanical punched-card processing machines. You set up the specialized gear-based machines for your job (payroll, overdue accounts, super-hard math equation), punched data onto ...
There have been many different kinds of I/O devices since the early years:
Indicator lights (often coloquially referred to as "blinkenlights")
Front panel toggle switches and pushbuttons
There is some more history available at the Computer terminal Wikipedia page.
I think the first robot to meet your criteria is Shakey the robot, developed between 1966 and 1972 at the Stanford Research Institute:
it could make decisions (it included a planner, and could break down complex commands into individual steps on its own);
its control logic was digital, running initially on a SDS-940 computer, later replaced by a PDP-10;
The Harwell Dekatron is an excellent example of this type of machine, not least because it was designed for simplicity and reliability rather than speed, and this in turn makes it easy to understand. This is a machine that operated entirely on decimal numbers, not on text.
It has a few buttons on its control panel, which are essentially used to start, stop ...
When your memory device is literally "a screen", you can just look and see what is held in memory.
(I think, but can't find a reference right now, that actually the display tube was a mirror of the actual storage tube - since the storage tube would have had a metal plate in front of it).
The video in the second link is worth watching to get a glimpse of ...
The first computer I built was a COSMAC ELF. It didn't have a keyboard or a monitor, no CLIUI/TUI/GUI. It was a full-fledged digital electronic computer.
It had 8 toggle switches and a pushbutton for input and a dual 7-segment LED display for output.
Before terminals, the most common method of interacting with large electronic tabulators, calculators, and computers (IBM mainframe, et.al.) was via punched cards for input, and line printer output for output. Some minicomputers used punched paper tape instead of punched cards.
The first terminals were mechanical teletypes (ASR-33, et.al.), with a keyboard ...
The word "intelligent" makes this a philosophical question. Here's one possible answer: scientists mapped out the brain of a worm and in 2014 they recreated it digitally and put it into a fully autonomous robot as part of the OpenWorm project. They didn't even have to teach it or give it any instructions.
Stephen's answer is already a good good one, at least when viewed by today's nomenclature. As usual, there's a grey area before such a clear point, one somewhere between the first, completely analogue versions of Elmer and Elsie or Wiener's Palomilla, and Shakey, the modern style robot.
Some good candidates may be "La Tortue Cybernetique" of 1951 by Paul-...