Hi - A good starting point would be to examine the map and text in http://www.chinalanguage.com

Stephen has given a fairly concise answer to this, your answer is basically there. One alternative is to be much more specific, and take natives of Beijing, for instance, who have been learning Mandarin in schools since year X. Research I have been involved with has had to say fluent speakers of X dialect. It's quite complicated, as pointed out above.

Japanese is more homogeneous only to an extent. Quite a few of the dialects are mutually intelligible, but they are actually quite different in some key areas. Other dialects, such as Aomori or Okinawa, can be very different from what one would consider to be "standard" Japanese.

Thank you all for your helpful reactions! Do you know if the gap between most people's L1 and the "standard language" used in media, at school, etc is bigger in Japan or in China?

I would be hesitant to conduct research on language comprehension without working with someone who was skilled in using that language.

But it's also important to note that you are dealing with literacy, not verbal fluency. It's particularly important with east Asian logographic like Chinese and Japanese (to a lesser extent), as the language when written is much more homogeneous than when spoken. Written Chinese in Guangdong is extremely similar to written Chinese in Beijing even if the pronunciation of what is written is extremely far from mutually intelligible. The homogeneity of written Chinese is one of the arguments used by those who claim that there is only one Chinese language with multiple dialects rather than multiple languages.

However, literacy is something that varies greatly, even among students attending the same university from the same town. In a nation of 1.3 billion people with substantial millions of people near the bottom of the socioeconomic system and substantial millions of people near the top of the socioeconomic system.

And while it is a tendency for many to consider China as part of a group with Japan (and Korea), China's population is about 10 times the population of Japan, and an area 25 times as large, though in both nations the population is largely contained in a relatively small portion of the nation. The languages of the two nations are vastly different. Many consider Japanese (and variations of it) to have no known relative, a language isolate. Japan also has one of the most complex writing systems of any language. This necessarily complicates study of written Japanese compared to written Chinese, or any other language.

On the other hand, the range in education in Japan is much smaller than in China, especially for relatively young people (like 35 or younger). Much of the Japanese population lives in the relatively small Kanto-Kansai strip from Tokyo to Osaka, and the economic range is much more restricted than in China, especially at the lower end.

While the Japanese writing system is partially based on the Chinese system, and the nations are relatively close in geographic terms, it makes no more sense to consider Japan and China as alternatives than China and Thailand or Japan and Vietnam.

I think that a decision on which language to use should be more closely aligned with the purpose of the experiment. The two languages offer what may be very useful contrasts with regard to studying logogram comprehension. Chinese is very much a logographic language. However, literacy in Japanese requires knowledge of logograms, 1-3 varieties of syllabic script, and alphabetic notation. You could exploit the complexity of Japanese writing or it could defeat the purpose of your experiment. But I think that if you don't already have a good understanding of the Japanese writing system (which doesn't necessarily require the ability to read Japanese), it might be best to exclude Japanese from your considerations.

In any case, you'll be limited by your access to experimental participants who are fluent in the languages you are investigating, and those who are bilingual at a sufficient level of literacy.

Thank you  for your extensive reaction, James! I will be researching into emotion recognition ability using audio-visual stimuli, though, so no worry about literacy.