Initially proposed independently by Benjamin Lee Whorf and Edward Sapir, the idea that language influences thought may initially seem counterintuitive. Clearly language and thought are related, we formulate and communicate thoughts using language, but it may seem that language merely expresses what is already there. While thought may not be wholly dependent on language as some have proposed, there is evidence to show that language does influence it.
One experiment which demonstrates such a connection between language and thought makes use of the grammatical concept of gender present in languages such as German and Spanish. These languages require that speakers refer to objects using the grammatical constructs of a certain gender. A believer of linguistic relativity might infer from this that the grammatical gender of an object influences perception of that object. To test this, speakers of Spanish and German were asked to describe the features of certain objects. When describing a key (which is feminine in Spanish and masculine in German), German speakers more frequently used words with masculine connotations such as jagged, hard, and heavy while Spanish speakers more frequently used words such as intricate, lovely, and shiny.
Linguistic relativity is also apparent in constructed languages such as computer programming languages. One example commonly used to illustrate this is Blub, a hypothetical programming language. Based on the availability of certain functions in programming languages, one might order them in terms of so called power. Blub is considered to be of intermediate power. A Blub programmer looking down the hierarchy at less powerful languages can see the functions missing from them and understands how Blub is more powerful. Looking up the continuum however, the Blub programmer is unable to see that he is looking up. Thinking in Blub much as one would think in a natural language, he would only perceive a language with seemingly bizarre constructs. For this reason, many programmers have advocated learning to use languages such as Lisp to become better programmers by broadening the way people can think about programs.
While these examples do not support the idea that language gives rise to thought, they do show that language influences thought in more subtle ways. The inherent properties of words can influence perception and even the understanding of concepts.
I can definitely agree with your example of gender being present in foreign languages. I have taken Spanish for five years and am constantly aware of masculine and feminine words. I didn't know that there had been studies done to see how German and Spanish speakers described the same word based on its gender, and this was very interesting to learn about. It would be interesting to find out how many other languages incorporate gender into their structures.