Earlier today I was listening to a German radio station, one that broadcasts a lot of public affairs and documentary material much like NPR in the States, and this time the story was about the mega-fauna extinction at the end of the Pleistocene era. Prominently featured were the La Brea tar pits and the Page Museum in Hancock Park. Talk about getting a story the long way around.
In the past, I’ve often noticed a peculiar thing when, in the course of a story, they interview someone who speaks English and have to translate that into German. Sometimes the person speaking English has a strong accent, because it’s a second language to them, but it’s the only language that they and the interviewer or translator both know. Or the person might be a native speaker, but have a strong non-North American accent that I’m not used to[sup]1[/sup]. When this happens, I find myself focusing more on the German, which actually sounds more familiar to me.
What was interesting about the story today, however, was that they were interviewing some curators at the Page Museum, as well as a paleontologist or two from UC Berkeley. In other words, they were evidently American native speakers with a noticeable California accent (strongly rhotic, among other things). In other words, it was the accent with which I’m most familiar. And this time I found myself focusing on the English speech. It was quite a bit more difficult than usual to understand the German translation, because I’d start listening to the English speaker, and just when the ball was rolling, so to speak, he or she’d be turned down and the German would come on–and then in the first second or two of adjusting to the language change, I’d miss part of the content.
So it seems to be true in my case that where English speakers are interviewed and translated on a German program, whether I focus on the English speaker or the German speaker is a function of how familiar I am with the English speaker’s dialect.
[sup]1[/sup]This wouldn’t be a standard BBC or Australian accent, but more something like Yorkshire, which we don’t hear much in the States. It can also be English spoken by a continental European from outside the German Sprachraum.
So what’s your point? 
When one is reasonably fluent in a second language, context often triggers the code switching IMHO. For example, I have a team that reports into me based in Taiwan. They are all Taiwanese. Actually, their English is fluent enough to at least not be torture. But we started off speaking Mandarin, and I’m sure we will go to our graves speaking Mandarin first with some English throw in. I’ve got opposite cases with other colleagues. Or my kids started out in Chinese, and then we switched to English, and now after a few years and living in the States, it’s tough to get them to speak to me in Chinese.
My point being that IMHO code switching is based a lot on context and what you’re used to.
I think that there is just something about hearing your native language with your native accent that is just much easier on your brain, and so if you’re listening to it, it makes it that much harder to “unpay” attention to it and focus someone speaking another language or with another accent.
This is just anecdotal, but I used to teach English phonetics in France, and for their midterm exam, one of the things the students had to do was to read a couple of paragraphs of English aloud. I would take notes on this, remarking on whether they pronounced their h’s, if they got the various th’s right, if they were stressing the correct syllable, and all that.
Now most of them spoke English with a British accent (usually RP), and some of them had really strong, but comprehensible French accents, and I paid attention fairly closely to them as I was listening. But there was one guy who had an American accent. And he was good. His speech was slow, it was halting, but it was undeniably absolutely correct American English. And even though I was paying careful attention to his speech, it seemed like I didn’t have to like I did for the others. It was somehow more directly targeted to my language-processing gray matter. It was like there’s some extra level of interpretation that your brain goes through to “decode” accents that simply wasn’t needed in this guy’s case.
I don’t know if there’s a scientific theory on this, but it seems that our two anecdotes point to the same conclusion, in that your native language with its native accent is somehow more primal in your brain, and that some additional effort of some sort is needed to listen to something else. So, when you listen to the German competing with the California English, the California English has an advantage in your comprehension that makes it harder to “switch off,” an advantage that doesn’t exist, or is weaker, with even other accents of English. Although, this all goes to hell if your native accent isn’t the Californian, and you’ve just been in California for a long time.
This is just my $/50. But I would be interested if there is a linguistics expert here who could shed some light on all this.