Is it possible to stop the conscious act of translating your native language to your second language or vice versa?
Many people have already commented that they certainly experience thinking in several languages. While this sort of phenomenal insight is interesting in itself, one key idea from cognitive science research is that introspection is not generally to be trusted when it comes to understanding cognitive processes (see for example the famous studies by Nisbett and Wilson). So both your conscious experience of thinking in your first language and my (or anyone else's) feeling of thinking in another language are not necessarily telling us much on how we think.
Is it possible to fully "think" in your second language?
I don't know but it's even possible to dispute the fact that you are thinking in a spoken language at all (see for example Jerry Fodor's “language of thought” hypothesis). Another view that is currently quite popular tries to account for abstract concepts in sensorial terms (cf. symbol grounding, some interpretations of “embodied cognition”, etc.) The idea is that when you are thinking about something, you are unconsciously manipulating sensory (visual, auditory…) representations, not linguistic ones (be it in a spoken language or some universal language of thought).
An example of an empirical result supporting this view is the “concept modality switch effect”: When asked to verify a property (e.g. “Can light flicker?”), participants are a little quicker if the previous question involved the same modality (in this case vision) rather than another one (say audition). This modality switch effect also exist when directly perceiving through two different modalities leading to the interpretation that to answer the question “Can light flicker?”, you are not merely retrieving some sort of symbolic knowledge (expressed in a language-dependent or language-independent manner) but actually visualizing/hearing/experiencing the relevant concept on a sensory level.
At the same time, you can also find theories purporting an influence in the other direction, in which the language one speaks changes the way one views the world all the way to pretty elementary perception (see Sapir-Whorf hypothesis). One area where this has been discussed and researched extensively is color perception, culminating in a big research project called the “World Color Survey” collecting data on color names in many cultures.
One line of research in this field uses differences between languages to track language influence on basic perceptual processes. For example some languages like Greek and Russian apparently have no word for “blue” but two words distinguishing different types of blue. The distinction between “dark blue” and “light blue” is obligatory when using these languages, you cannot say “blue” without specifying dark or light.
One intriguing result suggests that native Russian speakers are slightly quicker to distinguish two shades of blue that do not belong to the same linguistic categories whereas the effect does not exist for English speakers (Winawer, Witthoft, Frank, Wu, Wade & Boroditsky, 2007). Note that the task does not involve naming or remembering the color, only making a decision about two colors before you so it is pretty strong evidence that one does in fact “think” in a language (this is a complex question however, if you dig into this literature you will find that there is a lot of disagreement on various assumptions and methodological points regarding what is a basic color term, etc.).
Following the same line of inquiry, there is an empirical result that seem directly relevant to your question: Athanasopoulos (2009) used a color discrimination task with bilingual Greek/English participants (specifically Greek students in the UK). I don't remember all the details but he found that the effect varied depending on the level of acculturation (e.g. how long the participants spent in an English-speaking environment, how well they spoke English and at what age they learned it) and semantic availability of the relevant color terms (this was measured by asking people to list color names and counting how far down the list the different terms appeared). This suggests that, to the extent that language plays a role in perception, it is possible for a second language to have an influence but also that this is not an all or nothing proposition (i.e. the influence of the second language grows with increased exposure).
- Athanasopoulos, P. (2009). Cognitive representation of colour in bilinguals: The case of Greek blues. Bilingualism, 12 (1), 83-95.
- Winawer, J., Witthoft, N., Frank, M.C., Wu, L., Wade, A.R., & Boroditsky, L. (2007). Russian blues reveal effects of language on color discrimination. Proceedings of the National Academy of Sciences of the United States of America, 104 (19), 7780-7785.