What role should investigative facts play in the evaluation of scientific evidence?

Concern about contextual bias has led some authorities to recommend that forensic scientists know as little as possible about the facts of the underlying case when interpreting physical evidence; but concern about contextual ignorance has led other authorities to recommend, to the contrary, that forensic scientists know as much as possible in order to frame questions properly. This article recommends a case manager model that addresses both concerns. This article also responds to standard objections to the use of blind procedures in forensic science, explaining why contextual bias cannot be conquered through willpower; why use of domain-irrelevant contextual facts undermines the value of forensic evidence; how a well-known cognitive illusion (the 'introspection illusion') can mislead forensic scientists into thinking they can control their biases, when they cannot; and how a paradoxical feature of forensic inference (the 'criminalist's paradox') can mislead analysts into thinking they should rely on contextual facts, when they should not.