United States v. Semrau

United States Court of Appeals for the Sixth Circuit · 2012 · Evidence

693 F.3d 510 (6th Cir. 2012)

Updated August 22, 2026

EvidencefMRIlie detectionDaubertRule 702Rule 403expert testimonyscientific reliability

Facts

Dr. Semrau directed billing for psychiatric nursing-home services submitted to Medicare and Medicaid, and the government charged that claims were fraudulently upcoded from CPT code 90862 to 99312. To support his defense that he acted in good faith and lacked intent to defraud, he sought to introduce testimony from Dr. Steven Laken that fMRI testing showed he was generally truthful when he denied fraudulent intent. The testing was arranged by the defense without notifying the prosecution, used questions developed with defense counsel, and included multiple scans, one of which initially indicated deception before a later retest indicated that Semrau was not deceptive. Dr. Laken acknowledged limitations including lack of real-world validation, significant false-positive concerns, fatigue effects, and that the results did not establish truthfulness as to any specific statement.

Issue

Whether the district court abused its discretion by excluding expert testimony that fMRI lie-detection testing showed Dr. Semrau was generally truthful in denying intent to defraud. More broadly, the case asks how Rules 702 and 403 apply to novel lie-detection technology offered to bolster a criminal defendant's credibility.

Rule

Under Federal Rule of Evidence 702, novel scientific evidence is inadmissible when the proponent fails to show that the technique is sufficiently reliable in the relevant real-world setting and reliably applied to the facts of the case. Even if arguably relevant, lie-detection evidence may independently be excluded under Rule 403 when its probative value is substantially outweighed by dangers of misleading or confusing the jury, especially where the test was unilaterally arranged without government participation, raises credibility-bolstering concerns, and does not meaningfully establish truthfulness as to particular statements.

See the holding & full analysis

Create a free KwikCourt account to unlock the rest of this brief — and practice the case.

The court's holding and reasoning
Doctrine tests, pitfalls & exam hypotheticals
10 practice questions + 4 AI-graded essays on this case

Free sample · practice this case

Test yourself

One of 10 multiple-choice questions for this case. Pick an answer to see why.

In a wire-fraud prosecution in Detroit, defendant Nolan Pierce offers expert testimony from a neuroscientist that a brain-scan deception test showed he was truthful when he denied intending to mislead investors. The expert's published studies all involved college-age volunteers in low-stakes mock theft exercises, and the expert admits there are no validated error rates for use in actual criminal accusations years after the events in question.

Should the trial judge admit the testimony under Rule 702?

Explanation. Rule 702 requires not just testability and peer review, but a reliable fit between the principles and the use made of them in court. The majority held exclusion proper where the technique had not been adequately validated in real-world, high-stakes, long-term deception settings and lacked usable error rates for that setting. The court did not adopt a categorical ban on all such evidence.