In the article I will defend the view that cognitive science needs to use first- and second-person methods more systematically, as part of everyday research practice, if it wants to understand the human mind in its full scope. Neurophenomenological programme proposed by Varela as a remedy for the hard problem of consciousness (i.e. the problem of experience) does not solve it on the ontological level. Nevertheless, it represents a good starting point of how to tackle the phenomenon of experience in a more systematic, methodologically sound way. On the other hand, Varela’s criterion of phenomenological reduction as a necessary condition for systematic investigation of experience is too strong. Regardless of that and some other problems that research of experience faces (e.g. the problem of training, the question of what kind of participants we want to study), it is becoming clear that investigating experience seriously – from first- and second-person perspective – is a necessary step cognitive science must take. This holds especially when researching phenomena that involve consciousness and/or where differentiation between conscious and unconscious processing is crucial. Furthermore, gathering experiential data is essential for interpreting experimental results gained purely by quantitative methods – especially when we are implicitly or explicitly referring to experience in our conclusions and interpretations. To support these claims some examples from the broader area of decision making will be given (the effect of deliberation-without-attention, cognitive reflection test).