Both fonts fulfill the purpose for which they were designed.
Helvetica was built to live in a gridded world: the horizons and verticals of Swiss modernity (look again at R and see how space-consuming it is).
Arial was built to look like Grotesque 215, which is what swiss designers were using before Helvetica. Arial is supposed to be more human (read: humanist), and is for on-screen and on-paper --- not signage.
Designers prefer Helvetica because it is built to be used by designers; any combination of letters will pack neatly in space, both because of the letterforms themselves and the superior kerning.
Arial is meant to be used for less important, less permanent documents. It is more casual, and setting a whole page of text in it does not seem cold or callous.
All of which avoids the original question: "Does anyone know of reputable research and blind tests that back up the common designer sentiment against Arial?"
I don't think it's so much a matter of research or blind tests in this case. Like a regular user may not notice that a website loads .1 second faster because somebody optimized the database queries, a regular viewer will not be able to tell the difference between Arial and Helvetica. Designers, who are trained to be sensitive to those details will have an automatic response and be able to pinpoint what the reason is.
Alternatively, you could do the same thing with Arial and Helvetica. Show one to half of your visitors and the other to the rest then compare conversion results between the two sets.
There can be no such thing. Both fonts accomplish a purpose. Many like one because they feel it is easier to design with. Any reputable designer will instantly recognize both. They are two of the most used fonts in the world. There is no way to perform a blind test.
The closest thing to reputable researchers in design are designers. They spend their lives experimenting with fonts, and most prefer to use the one to the other. Some of them are following popular opinion mindlessly. Many aren't. Until some metric can be made for good design (fat chance. Also, note that readability is not the only important piece of this puzzle) no journal-worthy piece of science is going to come forward.
> Any reputable designer will instantly recognize both
No one is asking about "reputable designers". The tiny minority of people who can recognize hundreds of fonts at a glance are not interesting when it comes to this question.
There are certainly some objective metrics you can consider. Some that very quickly come to my mind:
1) Effect on reading speed
2) Effect on reading "endurance" (a little harder to define, but reasonable)
3) Effect on recall of text (prose)
4) Effect on recall in advertisements
5) Distance at which text of a given size becomes readable
There's also softer stuff along the lines of "does it make you feel warm and fuzzy", and you can still design experiments for that type of thing too.
So, I'm sorry, but I just don't believe you went you say you can't measure any of this.
I agree, readability can be tested. I am attempting to claim that readability is not the key factor. There have been font readability studies; but if designers were to follow them, all work would be black-on-white-large-x-height-minimal-serif-double-space-60-character-lines.
I am advocating seeing font as a tool designers use, not an end-user product. When I choose to develop a program using lisp instead of python (or vice-versa), I pick based on preference, knowing that I can produce a competant program with either --- though those programs will certainly differ somewhat --- not because of end-user useability studies of programs created with each.
I believe that equally readable, memorable, aesthetically pleasing design can be produced with either font; they are, after all, very similar. Designers find one more pleasant to use.
Helvetica was built to live in a gridded world: the horizons and verticals of Swiss modernity (look again at R and see how space-consuming it is).
Arial was built to look like Grotesque 215, which is what swiss designers were using before Helvetica. Arial is supposed to be more human (read: humanist), and is for on-screen and on-paper --- not signage.
Designers prefer Helvetica because it is built to be used by designers; any combination of letters will pack neatly in space, both because of the letterforms themselves and the superior kerning.
Arial is meant to be used for less important, less permanent documents. It is more casual, and setting a whole page of text in it does not seem cold or callous.