Streaming vs traditional

I think it is useful to distinguish between two kinds. The first is streaming from your own hardisk. This was the common way to do this: you rip (i.e. copy) your cds to you rown harddisk, or you buy downloads, and then access those from a device. The other is streaming from the internet, from services like Spotify, Tidal or Qobuz. If the latter are also full cd red book (losslessly compressed in e.g. FLAC) they will sound identical to cd's.  For now, Spotify only does 320 kbs (lightly) compressed, but Qobuz and Tidal also offer full red book cd quality. Whether you can hear the difference is a matter for debate (BBC research believe you cannot, and I am inclined to believe them). Anyhow, more bandwidth uses a lot more energy, and hence produces more pollution. However, in the footsteps of Qobuz and Tidal even Spotify is about to introduce full red book cd streams. My own preference is for streaming with a Chromecast Audio.

Spotify is apparently planning to follow Qobuz and move up to 16/44, the cd red book standard. That is not high resolution by most people's standard, but it is enough for me. I am not convinced that anything more is actually audible (there are claims, but no conclusive evidence).

For a blind test of MQA: http://archimago.blogspot.nl/2017/09/mqa-core-vs-hi-res-blind-test-part-ii.html

Of course I listen as well. However, sighted listening tests are methodologically deeply problematic because there is potentially so much expectation bias. The blind test that I linked to suggests that there is little or on sonic difference between high resolution pcm and mqa. Even those who thought they could identify the differences could not. In fact, the same has also been argued for differences between red book and higher resolutions.
This sighted test bias problem is not uncommon, and is well documented in the methodological literature. As a personal anecdote, I once participated in a blind amplifier test by Peter Walker. I thought I could identify differences, but I was completely wrong: like everybody else, I was no better than random. It was a good lesson that I have not forgotten. For a recent demo video on comparing amplifiers: https://vimeo.com/137001237
So the good take home news of all such blind tests is that there are few if any sonic differences between properly designed electronics (that may exclude sonically tweaked audiophile stuff). These days even quite cheap gear can be sonically perfect. Those who claim otherwise are often commercially interested (cables make people millionaires), or as consumers suffer from expectation bias delusion. The real differences in audio are with speakers and room interaction. And here, of course, the differences are easily audible, even if measurements help identify issues. Professionals have known this for ages.
As for my music choices. Well, I choose music, not demo discs.

Well, the comparison was between high resolution (i.e. 24/96) files and MQA, and (rightly or wrongly - there may be a discussion about the test design) the result was that there were no sonic differences. If true, this allows us to conclude that MQA is not better than Hi Res. However, this may be a correct conclusion simply because anything above cd red book is without additional sonic benefit, or because MQA is no better than Hi Res but both are better than CD.  That is not clear from this test.
I am not decided on the benefits of resolutions above cd red book. There have been blind tests and these do not seem to suggest a difference once the same source file is used (but downsampled). On the other hand, I have a number of BluRay opera discs and these are quite stunning. Is it because the format is superior, or because these recordings were simply done to high audiophile standards of dynamic range etc? Listening is not quite the simple thing it seems.

As I said, I am not sure higher rsolutions, either lossless or lossy compacted as in MQA are audibly better than red book. I think the jury is still out, but we shall see if someone can come up with decisive test data.
I don't do downloads for the reasons mentioned. I buy disks or I stream. The latter is a wonderful opportunity to get access to an amazing proportion of the world's music. I would like these streams to be of high quality, but at the same time better than perfect is obviously not necessary, and wastes scarce energy resources / pollutes the environment / adds to climate change. The scientists in the famous Phlips physics lab (Natlab) decided that 16/44 was the perfect sweet spot, with a resolution that was neither too low, nor unnecessarily high. Never again was such an outstanding team brought together in audio engineering. But if they were right, it was admittedly at the very edge of perfection, that is clear. These days, we would probably prefer to err a bit more on the safer side, i.e. 24/48.

To the best of my knowledge Meridian have never published any of their work in a scientific journal.

Wow. Even more voodoo science. I thought this forum already had its fair share.