Foundationalism about justification has historically enjoyed widespread acceptance among philosophers despite equally widespread disagreement about how foundational justification is possible. It is widely agreed that all knowledge must by justified by a foundation that does not stand in need of justification, but philosophers disagree on what could provide that foundation. Internalists, who look for justification in factors internal to rational agents, tend to agree that foundational justification is provided by seemings, or the way things seem to one to be. This view has most commonly gone by the name 'Phenomenal Dogmatism' although variations of it have been defended. Phenomenal dogmatism has been criticized for being too permissive with regard to the states it counts as able to confer foundational justification. In this paper I will consider one attempt, offered by Berit Brogaard, to revise phenomenal dogmatism in response to these criticisms. I will argue that Brogaard's revised view has significant problems of its own. Specifically, it does not account for problems arising from the possibility of cognitively penetrated perceptions. / Master of Arts / Many people take the way things seem (a ‘seeming’) to be good evidence for belief unless there is a good reason to think otherwise. This view is called ‘phenomenal dogmatism.’ In this paper, I consider a modified phenomenal dogmatist view proposed by Berit Brogaard. Brogaard claims that what she calls ‘evidential insensitivity’ is the mark of seemings that count as good evidence. I argue that her view both treats seemings as good evidence in cases where when it shouldn’t and also treats seemings as bad evidence in cases where it should treat them as good. One problem for Brogaard’s view comes from the possibility of cognitive penetration, which are cases where someone’s current beliefs influence the way things seem to them. Many philosophers have debated whether cognitive penetration can occur. Brogaard argues that many studies purporting to show that cognitive penetration can occur are methodologically flawed or have been misinterpreted. However, I cite studies that Brogaard has not considered and that are relevant to her proposal in particular.