I was thinking recently about the alignment (in the Dungeons & Dragons sense) of fictional scientists (see diagram above).
I was brought up on the Famous Five children’s stories by Enid Blyton. Perennially popular, even though flawed in certain ways, these novels star a rather grumpy scientist called Quentin (who had more than a little to do with my own desire to become a scientist). Quentin is certainly altruistic:
“‘These two men were parachuted down on to the island, to try and find out my secret,’ said her father. ‘I’ll tell you what my experiments are for, George—they are to find a way of replacing all coal, coke and oil—an idea to give the world all the heat and power it wants, and to do away with mines and miners.’
‘Good gracious!’ said George. ‘It would be one of the most wonderful things the world has ever known.’
‘Yes,’ said her father. ‘And I should give it to the whole world—it shall not be in the power of any one country, or collection of men. It shall be a gift to the whole of mankind—but, George, there are men who want my secret for themselves, so that they may make colossal fortunes out of it.’” (Enid Blyton, Five On Kirrin Island Again, 1947)
However, Quentin works for no organisation (barring some government consulting work) and draws no regular salary. He is clearly Chaotic Good.
Long before Quentin, Victor Frankenstein in Frankenstein (Mary Shelley, 1818) created his famous monster out of selfishness and hubris. However, he also desires to make things right, so Frankenstein seems to me Chaotic Neutral.
On the other hand, the experiments of Doctor Moreau in The Island of Doctor Moreau (H. G. Wells, 1896) mark him as Chaotic Evil. The same is true of the scientist Rotwang in the movie Metropolis (1927), who is the prototype of the evil “mad scientist” of many later films – in contrast to good “mad scientists” like Emmett “Doc” Brown in the Back to the Future movies (1985, 1989, 1990).
In all cases, however, there seems to be a bias towards portraying scientists as Chaotic. This is a little strange, because the organisational structures, processes, and rules governing science in the real world are better described as “ordered” or Lawful (in the Dungeons & Dragons sense). Perhaps chaotic characters are just more fun?
Not that everyone follows all the rules and procedures of course. When I take the What is your Scientific Alignment? test, my personal alignment comes out as Neutral Good.
Scientific Gems 
Doesn’t this reflect the serendipitous nature of science? Something Connie Willis seems to touch on in Bellwether, or Thomas Kuhn’s idea of paradigms. To make a significant discovery, you have to ignore conventions and rules.
There may be something in that (and I’m a big fan of Bellwether). It’s still true, however, that most scientists work within a paradigm. And those who step too far outside the paradigms often produce junk science, because checking and verifying results is a communal activity.
But even inside the scientific and mathematical communities, we idolise mavericks like Paul Erdős (perhaps because, deep down, our dream is to be one of the successful mavericks).
From the outside, of course, the mavericks and the unusual people are far more interesting to write about. This is why everyone talks about Alan Turing solving the Entscheidungsproblem, and almost nobody talks about Alonzo Church, who solved it first.
And What is your Scientific Alignment?
I agree with your comments about result checking being communal, and being too maverick is a problem. Scientists do have to be deeply connected and informed by current paradigms to advance on them, because it is in the areas where the paradigms break down that advances can be made.
I was thinking that from my personal interactions with other scientists, they do tend to be a bit odd compared with people in other circles I interact with, so I would hazard a supposition that the trope is drawing on a real phenomenon.
I come out as neutral good also… although I had a bit of trouble with some of the questions. For instance, I agree that the peer review process is better than nothing, but there are quite a few instances showing it doesn’t solve all problems, But, it’s an imperfect world – we do the best we can.