Human cognition is still a puzzling issue in research and its appropriate modeling. It depends on how the brain behaves at that particular instance and identifies and responds to a signal among myriads of noises that are present in the surroundings (called external noise) as well as in the neurons themselves (called internal noise). Thus it is not surprising to assume that the functionality consists of various uncertainties, possibly a mixture of aleatory and epistemic uncertainties. It is also possible that a complicated pathway consisting of both types of uncertainties in continuum play a major role in human cognition. The ability to predict the outcome of future events is, arguably, the most universal and significant of all global brain functions. The ability to anticipate the outcome of a given action depends on sensory stimuli from the outside world and previously learned experience or inherited instincts. So, one needs to formulate a theory of inference using prior knowledge for decision-making and judgment. Typically, Bayesian models of inference are used to solve such problems involving probabilistic frameworks. However, recent empirical findings in human judgment suggest that a reformulation of Hierarchical Bayesian theory of inference under this set-up or a more general probabilistic framework based approach like quantum probability would be more plausible than a Bayesian model or the standard probability theory. However, as the framework of quantum probability is an abstract one needs to study the context dependence so as understand the new empirical evidences in cognitive domain.