This paper deals with the argumentative biases Conspiracy Theories typically suffer from and pursues two goals: the identification of recurring argumentative and rhetorical features of conspiracy theories, which translates into an attempt to elaborate their argumentative profile ; the elaboration of a cognitively-grounded account of CTs in terms of their persuasiveness. To fulfil goal, I examine online instances of different cases of CTs. Building on the general rhetorical features of CTs identified by Byford, I elaborate a first argumentative profile surveying (...) types of arguments and argument schemes CTs are likely to rely on to specify how these “crippled epistemologies” are argumentatively biased. To fulfil goal, I use the Context Selection Constraint model, originally designed to capture deception. I examine the cognitive counterpart of the – most often fallacious – arguments found in CTs and advance cognitively motivated reasons why CTs have some prospect of being argumentatively effective. This provides an account of why CTs, while biased, can manage to persuade, that is, precisely by downplaying – or obfuscating – their bias. (shrink)