Polygraph tests have been used to detect deception for almost a century. Yet for almost as long, the validity of these tests has been criticized. Over the last decade, the use of brain imaging - most notably fMRI - for the detection of deception has attracted increased attention. The expectation is that fMRI can overcome - at least some of - the shortcomings of the polygraph. In this review, we discuss whether this expectation is warranted. Based on our review of the empirical evidence, we argue that fMRI deception research has boosted the theory development of deception. But for practical purposes, fMRI research has thus far done surprisingly little to solve or circumvent the problems than have been associated with deception research for ages.