Many cultures have stories that are told as fiction as well as those that are told to be believed - generally. In English, the two forms are referred to as folktales or fairytales on the one hand and legends on the other. European cultures echo this dichotomy, and there is evidence of these two types of narratives having existed for millennia.

The folktales were not usually intended as children's stories - they were the novels of the folk, and there is clear evidence of storytellers taking several nights to recount them - with adult content, told after the children went to bed.

When looking at narratives in ancient or medieval sources, it is often - but not always - apparent which is which. Indeed, sometimes stories that were told as folktales in one context could drift over into the realm of legends in another. In addition, folklorists assert that legends are "generally told to be believed," leaving room for some of these narratives to slip into the realm of fiction - at least for some people in some contexts.

All this does, in fact, cause a certain amount of blurring of the lines, but these are the boundaries that can be applied, and because they consistently appeared in European cultures, it is often possible to recognize which is which.

Legends about King Arthur were often told to be believed: folklorists working in Wales, for example, were able to document believing in these stories. The stories about Jesus also fit the criteria to be called legends, but there is an important point to make here: just because we refer to a narrative as a legend does not mean that it is told to be believed AND it is false. Folklorists do not pass judgement on survival of the soul in an afterlife when looking at legends about ghosts. They are stories that the folklorist evaluates, but the evaluation does not begin with the assertion, "well, this is clearly BS, so it is folklore." Rather the folklorists begins with the assertion of "people believe this and they tell this story, so we can regard it as a legend - whether or not it is true."

I hope this make sense and is helpful. Let me know if you have questions.

I am more of a folklore student than history, but I also study Celtic Civ, so I can give you my perspective on this at least, but the answer will no doubt vary depending on the culture and time being studied.

One important thing to understand is that the two categories you present are not necessarily a binary. It's not uncommon at all for people to only partly believe in a story or supernatural element of the world as they perceive it. Most of the people in an area might believe that a certain practise wards off disease or had luck etc etc, but even those who supposedly don't believe in "that kind of thing" may still participate in the practise on part or whole "just in case" or because they feel compelled to through societal obligation or habit.

An example of this are "fairy forts" and "fairy trees" in Ireland where most of my study is focused. (The word fairy is problematic in this context but that's beside the point.) These places and plants are associated with supposedly supernatural forces, and are therefore usually left alone, and stories are told about those who don't. Now, in recent history, if you were to ask many people of they believed that otherworldly hosts would come take them away if they trespassed there, they would say no. But they still stay away. Because if you're told growing up over and over that something is dangerous, you're probably going to have some level of fear of it. Whether or not you or the person telling you believed the "story" part of it.

So this complicates your question. Belief isn't a yes or no question, belief can he denied but practised. Now as for medieval material, that's even harder because knowing to what extent the material reflects the real world is always a question. I'm a little less equipped to answer this part of the question, but from what I can tell the basic rules of analysing the source, comparing it to other textual sources and comparing it to archaeology where extant seems to he the rule.

There is still the same issue as before with "pseudo histories" lying somewhere in between supposedly real history and storytelling, but generally as far as I can tell from what I can tell, at least with the material I study, the practise of taking medieval material as a direct reflection of belief at the time or before is no longer in vogue. If it is a surviving textual source, it is probably more literature than a directly transcribed oral tale, at least in medieval Irish material, and so is probably an edited or compiled work based on existing stories, or an original text pulling from beliefs and tales of the time just as authors do now.

In that way it's best not to read these tales as what people believed so much as tales that include elements of the society that produced them. If a certain location is repeatedly cited as a place that supernatural characters appear, it's reasonable to believe there may have been some genuine belief about that place, if a certain item is used to perform some supernatural task in different texts, then clearly there's something there.

I hope that was helpful to some extent, but basically as far as I can tell, there won't always be a clear distinction between what is believed or not, and medieval texts are often better used as insights into the time and place that created them rather than as direct records of belief.

You may find very interesting to read previous replies about Jesus. How do we know Jesus actually existed? Why also atheist historians don’t think Jesus is just a myth?

You can find many interesting threads on the wiki

https://www.reddit.com/r/AskHistorians/wiki/faq/religion#wiki_christianity