The very nature of thought has often been understood to be constitutively related to language use. This view, lingualism, falls foul of David Hume’s animal test, according to which the plausibility of a theory about the mind is, in part, determined by its implications regarding nonhuman animals. Lingualism about thought does not pass Hume’s animal test because it implies that animals which do not make use of language are ipso facto unable to think. A theory that meets the demands of the animal test is Jerry Fodor’s Language of Thought hypothesis (LOTH), according to which thought is not constitutively connected to any natural language such as English or Italian, it rather bears the form of mental representations, which are structured according to the principles of a mental language, which is innate. However, the LOTH also faces a problem regarding animal thoughts, namely the problem of indeterminacy in translating animal thoughts. Since, according to the LOTH, animal thoughts bear a language-like sentential format, namely the format of the Language of Thought, they should in principle be translatable into sentences of some natural language. But, as both empirical and theoretical research suggest, there are strong reasons to believe that insurmountable indeterminacies as regards the interpretation of animal thought into sentential formats would strongly distort the contents of the thoughts. This problem of indeterminacy can be tackled if we distinguish different formats of thought: sentential formats and non-sentential formats, such as cartographic thought. Motivated by the problems animal minds pose for classical accounts of the mind, this PhD project will elaborate how non-sentential formats of thought help to explain both similarities and differences in cognitive abilities between humans and nonhuman animals. This will help us to better understand the aspects of our thoughts that we share with animals and what makes some of our thoughts distinctly human.