"Wa gwaan, people?" :) [clue: Jamaican patois]
(I know, this is not just an accent, it is patois but it helps to make my point about exposure and experience.) I heard this in a TV series with Jamaican characters who kept saying it over and over. It finally hit me: What's going on. :)
Leaving aside comedy as mentioned in the question, understanding an accent different from one's own is a matter of two main factors:
Exposure and Experience. And also, bear in mind that vocabulary and phrases will probably also be factor. A third factor can be what I call "the talented ear". Just like with music, some people can hear and reproduce an accent at first hearing. Others might be accent deaf (like tone deaf) and have a very hard time understanding an unfamiliar accent.
Bear in mind, there are many, many accents in English and one must not forget foreign accents with a typical sound.
This phenomenon about understanding accents is quite difficult to study scientifically. One would have to do something like this:
- Set up a native English control group from a geographical area to ensure they all speak the same variety of English.
- Read or play a text for them in an accent in English that is not their own.
- Have them answer a questionnaire about that experience.
Even so, I don't know how scientifically valid it would then be to make broad sweeping statements about "understanding other accents" on the basis of that type of exercise. Many of these types of experiments would have to be conducted, and cross-validated. The endeavor could be nightmarish as finding statistical relevance could be very difficult. I can imagine a situation where some Group A of English speakers could barely understand some other accent, whereas some other group, Group B of English speakers, might understand the other accent better. One's starting point would have to be clearly defined. In other words, for example, can a Londoner understand an American southern drawl more easily than a Liverpudlian? Drawing on comparative studies of accents might help here prior to setting up the survey.
In those three statements, one could replace the word English with any other language, and a variation of that accent. Think of a country like India. There are several hundred languages spoken there. This is an interesting statistic in that regard:
As per the 2011 Census of India, languages by highest number of speakers are as follows: Hindi, Bengali, Marathi, Telugu, Tamil, Gujarati, Urdu, Kannada, Odia, Malayalam. [Wikipedia]. Think of running an accent survey there...:)
The exposure criteria is crucial. Although one speaker may understand an accent, another may not. One comes to understand accents through one's lived experience of them. However, it may also be true that one can understand a new accent the first time one hears it. Someone who is used to different accents in their own language will probably understand one that is new to them more easily than someone with little or no exposure to other accents.
It's interesting to note that the BBC (UK) and PBS (the U.S.) sometimes provide closed captioning for some speakers in English they are reporting on in their news reports or video clips. Sometimes, these English speakers are from Africa or India, for example, where there are some very strong lilts (which can be quite beautiful, in fact.)
Anyway, accents make the world go round and so does comedy. The greater the exposure to different accents through travel, media or personal interactions, the greater the ability to understand them, as a general proposition.
A non-native speaker of English can obviously master English at a high level. When one does master a language at a high level, understanding more than one regional accent in that language should not be that much of a problem, at least to a certain extent but maybe not every nuance.