Foreign Language Grammatical Errors as a Native Language Identifier

Sounds like the title of an impressive paper to be submitted to some Linguistics conference.

Actually it's just the best way to describe an interesting thought I stumbled upon. I'm reading through a number of papers for my research and the authors come from many different cultures. There are a vast number of European researchers in my chosen area of expertise, and many papers come out of Germany, Switzerland, and France. A large number of papers also come from various Asian sources or authors. As I was reading through a paper titled "Content-Based Retrieval of MP3 Music Objects" I was getting the nagging suspicion that the author was Asian and learned English as a second language based solely on the "speech" pattern and the consistent type of grammatical errors being made. I checked, and sure enough, the authors are Chih-Chin Liu and Po-Jun Tsai from Taiwan.

For examples of what I'm talking about, take these sentences:

'This is because in this time interval, the note corresponding to the word "empty" becomes very loudly and its pitch falls into these two subbands.'

'Approximately matching ability is strongly required in a multimedia database system.'

'These structural audio information can be used to support content-based retrieval.'

I'm not picking on anyone here, it's just an observation that I found interesting. Coming from specific native languages will cause a greater likelihood of specific types of mistakes in foreign language. I would hope this interesting phenomenon has been studied before, perhaps Katya can enlighten us further on the matter.