In case you are unaware, the question of how much of our thinking is language-constrained has been extensively studied in linguistics and cognitive science. http://en.wikipedia.org/wiki/Sapir-Whorf_hypothesis is a good starting point. Anecdotal evidence is great, but you should also look into the studies.
My personal perspective: I am natively trilingual, as an Indian person. (someone else mentioned this as well.) My two Indian languages have fallen to near-complete disuse, to the point that I'm only borderline fluent in them. Nevertheless, growing up with two or more languages forces a child to seek common denominators, and thus think in terms of imagery.
This is definitely the case for me. As soon as I focused on the word "empathy" for a few seconds, I actually had two different images come to mind. These were both exemplars, as in, specific instances of people in my life showing empathy, one from recent memory and one from long ago. That's the trick with imagery -- it can often be crude, and only cover a special case of the concept, but it still does the job much better than words. You mention precision. Precision can in fact be a disadvantage, given the inherent ambiguity of human thought.
This is learnable. It gets harder as you age, but never impossible. To improve thought imagery, try this. Buy a ginormous whiteboard and cover one wall of your room/office with it. Draw everything. You might initially have to struggle with coming up with any sort of image. Your perfectionism, and your poor perception of your drawing skills, if you suffer from those, might hinder you. Try and lose your inhibitions. When you're at your computer, keep looking at the whiteboard once in a while. You want to get to the point where if there's a concept you're working on over a few days, every time you think of that concept, you should immediately be able to see the corresponding picture in your mind's eye. Also keep reams of paper around. Again, draw everything. I'll even start drawing on napkins if a thought comes into my head that I need to draw and I can't find paper. Hopefully, you will eventually be able to 'draw' effortlessly in your head.
Here's a solution for your "lookup table" problem. This is a technique that is used in teaching accents, but I have also used it in learning new languages. Pick a character. It could be a friend, or a character from a TV show, just anyone who speaks Spanish (preferably exclusively). When you're trying to speak Spanish, be that person. Imagine you're them, to the extent possible. It will make it a lot easier to push the English out of your head.
Finally, subvocalization (i.e, reciting sentences to yourself). Also (un)-learnable. Start gliding your eyes over the text rapidly. At first, you won't understand anything. Try the same sentence again. Force yourself not to say the words. If it's too hard, go fast enough that you can't possibly say the words. Keep doing that until you derive at least some meaning from the sentence. Start from there, and hopefully you'll get better slowly.
These are all things I've used to various degrees. Thinking in English is not a big problem for me, but more imagery helps, no matter what. YMMV, but hope this all helps!
I think the Sapir-Whorf hypothesis is a pretty imperfect fit for the question. Even the weaker form doesn't really gel well. The question seems more do certain modes of thought make you tend to think of certain languages, and not so much "do certain languages impact behavior".
Basically, its generally formulated in the opposite manner, and like many things, cause and effect can not usually be reversed and still have the reasoning hold ;)
Also, I grew up effectively trilingual as well. I simply associate the different languages with different areas of thought. Math/science is English, food/home-y things is Chinese (and er, well, Taiwanese+Japanese also food/home), and I really don't think in terms of imagery. So I think you're definitely in anecdote territory. (BTW: I actually count in Chinese, because its a tad more uniform and the words are shorter, but operators I use English.)
I wonder if language learning methods will vary from person to person, because most of your advice totally disagrees with my experiences learning Spanish later in life. I learnt through rote memorization, but followed up with a conversation partner, so, to continue the lookup table analogy, I simply used the lookup table until it became so commonly used that it stayed in cache and not memory. Then over time, segments were pulled out and their structure transformed (length of cache value increased, keys switched to a general emotion), until more and more of it was "fluid".
(The other possibility, which I'm very inclined to believe, is that describing your own learning process is a crap shoot. You find analogies you're pleased with, and selection bias takes over.)
I take a medium-strong form of S-W. It makes sense in the connectionist model (which I also take a medium-strong interpretation).
The strongest form says that concepts in one language are completely alien to another, such that if you don't know the concepts in one language, you cannot perceive them. This is obviously false.
However, as related concepts are evoked by spread-activation, it is completely plausible that one language allows more efficient retrieval of certain concepts, such that very strong biases result.
The famous example of the many Inuit words for snow comes to mind. It is true that a non-Inuit speaker is able to discern between different kinds of snow, but the language guides the mind in terms of where to look for details. Without this guide, it is easy to get lost and confused.
Here's the familiar example for hackers: code in C/Basic/Pascal for a while, and switch to Python/Ruby (or vice versa). After you get used to the target language, and look back at the code right after the switch, you'll notice that a lot of programming constructs are out-of-place and "counterintuitive." This is, so to speak, Sapir-Whorf "at work."
This is also why when LISPers say that you should learn lisp, because it will make you a better programmer, they are correct. :)
The comparison with programming languages is extremely misleading, because, unlike natural languages, programming languages do differ in the expressive power. An algorithm coded in Haskell or Lisp may well be 5x less code than an equivalent thing in C or Java. On the other hand, when a text is translated from one natural language to another the result is typically of the same length and typically there's even an almost one-to-one mapping between individual sentences.
To anyone interested in the relations between the language and thinking, I recommend an excellent book "The Stuff of Thought" by Steven Pinker. One chapter is devoted to the Sapir-Whorf hypothesis, more specifically, to debunking it.
Two things: again, I don't espouse the strongest form of Sapir Whorf. The interpretation of differences in speed of access to concepts, based on spread-activation, however, is solid. When I learned Sapir Whorf we were given the original interpretation and a brief of how it has been reinterpreted more recently -- but it still goes by the name of Sapir Whorf. I am not making a case for the possibility of expression. I am make a case for the salience of expressions in different languages. It should be easy to show this: take two languages and find two words of identical meaning, and measure the response speed to comprehension of these words. All other things equal, the higher-frequency word will be accessed faster. Hence, if one culture says Apple more than Orange, Apple is more salient in the former culture. You will then expect more efficient access to Apple-related concepts.
Second, what makes you think that natural languages do not differ in expressive power? Certain things are more easily said in one language than another, just as some constructs may be easier to represent in one language or another, although they may be able to create the same results.
I also don't know where you get the "typically the same length" from. Stephen Pinker wrote about the relative length of English in one of his books. I don't remember which he was comparing to, but English is longer. Pinker explained this along the lines of English being more "fault tolerant" due to more "redundancy" (my words).
This is probably more a function of the fact that languages change over time, so if a common task is hard to talk about, new forms will appear that make it easier. Programming languages see very little change (and even rarer are ambiguity or "non backwards compatible" changes introduced.)
Stuff like this is why I became interested in PL theory in the first place actually.
My fault for not confirming the veracity of that story in particular, but I was using it as an illustration of accessibility to concepts in different language, independent of its truth. I just supposed that this story is widely known, but I could have (perhaps should have) easily used another example. There are several words in Chinese that describe taste without direct English translations that I know of (it is harder to sustain these ideas nowadays, given the quickness of word-borrowing like "umami"). One of them is "fish stink," which really isn't fish stink. A strong S-W will say the concept is inaccessible to the non-Chinese speaker, and, of course, that isn't true.
In fact you don't even need to cross language boundaries to see the same effect: layman versus technical terminology reveals analogous differences in "perception." Me not knowing the name of industry colors does not mean I cannot perceive these colors (let's assume a single-sex population), but I will gladly group together large groups of colors into single words.
I suggest you read the wikipedia article. Theres a pretty good summary of its current status. "well, dub it has some effect, but it doesn't make certain thought impossible"
A better example is that some languages have more words for color, and thus people are able to differentiate between slightly different more shades. Though I'd argue thats more a function of training than the mere existence of more word.
Strangely, for me, in imagining empathy, all that comes to mind for me, is an abstract understanding of it. I can "feel" its meanings, almost in the sense that the meaning flows through my fingertips, but I don't experience imagery or episodes or sounds.
Similarly, when I need to use an abstract word, I translate these feelings into words. Often times, possibly as a result, I come up with strange words that fit the context but in an unconventional manner.
While this probably sounds a bit strange, babbling in freefrom gibberish and stretching your mouth / jaw muscles can also help to relax sub-vocalization. It seems to relax an underlying vocal fidgetiness the same way a good stretch/workout does for the rest of the body.
In other words, additionally approaching sub-vocalization as a physical phenomenon might be helpful.
Steven Pinker pretty convincingly demolished S-W (for me) in "The Stuff of Thought". He argues that there is an abstract linguistic framework upon which the concepts in a language are built, things like content-locatives, container-locatives, causatives, and much more. Recommended book.
My personal perspective: I am natively trilingual, as an Indian person. (someone else mentioned this as well.) My two Indian languages have fallen to near-complete disuse, to the point that I'm only borderline fluent in them. Nevertheless, growing up with two or more languages forces a child to seek common denominators, and thus think in terms of imagery.
This is definitely the case for me. As soon as I focused on the word "empathy" for a few seconds, I actually had two different images come to mind. These were both exemplars, as in, specific instances of people in my life showing empathy, one from recent memory and one from long ago. That's the trick with imagery -- it can often be crude, and only cover a special case of the concept, but it still does the job much better than words. You mention precision. Precision can in fact be a disadvantage, given the inherent ambiguity of human thought.
This is learnable. It gets harder as you age, but never impossible. To improve thought imagery, try this. Buy a ginormous whiteboard and cover one wall of your room/office with it. Draw everything. You might initially have to struggle with coming up with any sort of image. Your perfectionism, and your poor perception of your drawing skills, if you suffer from those, might hinder you. Try and lose your inhibitions. When you're at your computer, keep looking at the whiteboard once in a while. You want to get to the point where if there's a concept you're working on over a few days, every time you think of that concept, you should immediately be able to see the corresponding picture in your mind's eye. Also keep reams of paper around. Again, draw everything. I'll even start drawing on napkins if a thought comes into my head that I need to draw and I can't find paper. Hopefully, you will eventually be able to 'draw' effortlessly in your head.
Here's a solution for your "lookup table" problem. This is a technique that is used in teaching accents, but I have also used it in learning new languages. Pick a character. It could be a friend, or a character from a TV show, just anyone who speaks Spanish (preferably exclusively). When you're trying to speak Spanish, be that person. Imagine you're them, to the extent possible. It will make it a lot easier to push the English out of your head.
Finally, subvocalization (i.e, reciting sentences to yourself). Also (un)-learnable. Start gliding your eyes over the text rapidly. At first, you won't understand anything. Try the same sentence again. Force yourself not to say the words. If it's too hard, go fast enough that you can't possibly say the words. Keep doing that until you derive at least some meaning from the sentence. Start from there, and hopefully you'll get better slowly.
These are all things I've used to various degrees. Thinking in English is not a big problem for me, but more imagery helps, no matter what. YMMV, but hope this all helps!