I would like to know if its possible to test blind XSS via a cookie value. That is I send the server some cookie that has a blind XSS payload. Or is blind XSS only exploitable if a human interacts with my payload for instance through a web form?
Put simply, blind XSS is a form of XSS where the adversary cannot access the endpoint where the payload is stored/executed. The important property of XSS that is relevant to your question is that the payload consists of client-side code.This means in order for the payload to execute, a client must render the DOM containing the malicious payload. This so called "client" could be an unsuspecting user or might go as far as being a headless browser triggering the code.
Based on personal experience as a bug bounty hunter, blind XSS most commonly occurs via submission forms where the payload is stored in a dashboard which an employee can then access.  That being said, there have been plenty of cases where the payload is injected via the
Referer, etc. headers and then stored in a fancy HTML log somewhere on the company's infrastructure. At the end of the day, if the application is storing data unescaped somewhere, blind XSS becomes a plausible attack vector. Maybe placing an XSS payload in your cookie might trigger an error that is stored somewhere unescaped. ¯\_(ツ)_/¯
On a side note, if you are planning on hunting for blind XSS vulnerabilities in the context of a bug bounty program, for instance, I would highly recommend using XSSHunter and experimenting a bit. If someone triggers your payload, you will receive a pingback from XSSHunter notifying you where your payload ended up.