Thought this might be interesting and it hasn't been posted here yet.

In 2014, Jeffrey R Brown wanted to resolve a debate over what makes an IAL easier to learn: Simplicity or Familiarity (putting aside other factors e.g. socio-economics, aesthetics, etc). To investigate this he asked volunteers to learn a conlang that was both simple and unfamiliar, then tested them on their translation accuracy.

If Simplicity was the most important factor, you'd expect everyone to be equally accurate at the conlang. However there was a lot of variation in accuracy, which correlated with the learner's linguistics knowledge (e.g. number of natural languages spoken, number of conlangs constructed). Basically, more languages you're exposed to -> greater accuracy, so simplicity is less important.

By Simplicity, he means a regular grammar with few rules (admitting that 'simplicity' can only be measured very crudely). Since the volunteers had no knowledge of Polynesian languages, the 'unfamiliar' conlang was based on modern Hawaiian (Ma'ahali - grammar and lexicon).

​

I love that he actually did a study, though I have some criticisms:

• A control group could help measure how much variation you should expect. For example, if some of the volunteers were tested in Esperanto and the variation was the same, then familiarity isn't related. (EDIT: I say Esperanto because he designed Ma'ahali to be roughly as simple as it).
• When we talk about the familiarity of an auxlang/IAL, we don't mean language-learning ability, which is what the study ends up measuring, IMO. The author does specify that there's no correlation with level of formal linguistics study, but I believe that creating a detailed conlang gives you an advantage in language-learning, which could affect these results. The volunteers were drawn mostly from conlang communities.
• Volunteers mentioned that the learning material was terse and the author admits that it was written for linguists - not as a primer. That could've increased the effect above.

Despite that, this is a really valuable study and it's amazing that he actually went out and tested these theories.