On March 12, 2026, a diverse group of voices filed amicus curiae briefs in support of Charles Flores (pictured), a Texas death-sentenced prisoner, urging the U.S. Supreme Court to review his case. Mr. Flores has spent more than 25 years on death row for a murder he maintains he did not commit. His conviction relied on the testimony of a neighbor who identified him — for the first time, at trial — only after being hypnotized by police. The briefs were filed by a coalition including the entertainers Penn & Teller, the American Psychological Association, crime survivor Jennifer Thompson, the Texas Defender Service, and exoneree Christopher Scott, all of whom offer different perspectives as to why Mr. Flores’ case presents grave constitutional questions. The coalition is united in the argument that “investigative hypnosis” does not unlock hidden memories, but rather distorts them, leaving an inflated sense of certainty about memories that may be partially or entirely fabricated. Mr. Flores’ case sits at the heart of a broader issue of mistaken witness identification across the criminal legal system in the U.S. that has sent hundreds of innocent people to prison and may potentially result in his execution.
“[Charles Flores] has not only undermined confidence in the conviction; he has presented persuasive evidence, grounded in contemporary science, that is affirmative support for an innocence finding building on the post-conviction record establishing the complete unreliability of testimony from a hypnotized witness (a record that was entirely ignored).”
In a petition for a writ of certiorari filed in February 2026, Mr. Flores asked the Supreme Court to consider whether the Texas Court of Criminal Appeals’ consistent refusal to apply state laws meant to prevent the execution of innocent people violates his constitutional right to due process. Among the briefs filed in support of Mr. Flores is one from Penn & Teller—the world-famous duo who have spent five decades performing illusions, debunking pseudoscience, and studying the cognitive science behind perception. Self-described “honest liars,” they filed their brief expressly “to explain how many of the cognitive manipulations they use to trick audiences’ perception of truth are on full display” in the hypnosis session that produced the only identification ever made against Mr. Flores. Their argument centers on how memory actually works, and how law enforcement exploited it in the case of Mr. Flores. As Penn & Teller explain, “all of your memories are copies of copies of copies,” and those blurred copies, they argue, create the same openings for suggestion that they exploit nightly on stage. The officer who hypnotized the witness in Mr. Flores’ cases told her explicitly that her mind would work like a documentary camera, a premise Penn & Teller identify as “one of the biggest lies about hypnosis[.]”
The officer also assured the witness she would “remember more” afterward, priming her to later “remember” a face she had repeatedly failed to identify. Just as a magician uses subtle verbal cues to lead an audience to remember a card shuffle that never happened, the officer repeatedly suggested the suspect had short or shaved hair, gradually reorienting the witness’ memory toward a description that matched Mr. Flores. Penn & Teller conclude that “there is something fundamentally amiss in the justice system if flim-flam like investigative hypnosis can be used by law enforcement to reconfigure the gap-laden memory of a key witness in a capital prosecution,” and that it is “not just ironic but illogical and unjust” that Mr. Flores’ case inspire a Texas law banning investigative hypnosis in criminal proceedings, yet he remains on death row denied each opportunity to challenge its use in his own case.
The other amici reinforce this picture from their own perspectives. The American Psychological Association, representing approximately 174,000 members and affiliates, explains that initial identifications are the most reliable, and that a witness’ failure to identify someone early on, as happened here, is “particularly probative of innocence,” while in-court identifications made after repeated exposure to a suspect’s image are the least reliable. Jennifer Thompson, who spent years believing she correctly identified the man who raped her before DNA evidence exonerated the man she had wrongfully identified, urges the Court to ensure Mr. Flores is not executed based on the same kinds of suggestive procedures that corrupted her own memory. The Texas Defender Service (TDS) presents data showing that the Texas Court of Criminal Appeals (TCCA) has “systematically frustrated” the state’s own “junk science writ,” a law designed for exactly this kind of evidence, by imposing procedural barriers that effectively foreclose review of capital cases. TDS notes that more than a dozen death-sentenced individuals have sought relief under this law, Article 11.073, yet the TCCA has denied relief each time. Dallas County exoneree Christopher Scott argues that Mr. Flores’ prosecution bears the same hallmarks that now define the wrongful conviction playbook: unreliable forensic science, mistaken eyewitness identification, and prosecutorial misconduct.
Mr. Flores was convicted and sentenced to death in 1999 for the 1998 robbery and murder of Elizabeth “Betty” Black in her Texas home. He was convicted based on the testimony of Jill Barganier, one of Mrs. Black’s neighbors, who only identified Mr. Flores after being hypnotized by police — 13 months after the crime occurred. No DNA or physical evidence ties Mr. Flores to the crime. The jury did not know that immediately following the crime, Ms. Barganier described seeing two individuals leaving a distinct VW Bug and going into Mrs. Black’s garage, neither of whom look like Mr. Flores, and she failed to identify him in a photo line-up at that time. The first time she identified Mr. Flores was during trial, after she had gone through the investigative hypnosis and been exposed to Mr. Flores’ photo on multiple occasions.
At the time of the crime, Ms. Barganier identified Richard Childs out of a six-photo array. She was unable to identify the second individual, even after being shown an image of Mr. Flores, and men who actually matched her description of the second individual (a white man with long hair). Under investigative hypnosis by a police officer with no hypnosis experience, she repeated her identification of the second man, characterizing both individuals as white men with similar builds and long hair. Mr. Flores did not match her description — he is Hispanic and wore his hair shaved short — contradicting the identifying factors brought to law enforcement’s attention. Following the hypnosis session, Ms. Barganier again failed to identify Mr. Flores in a photo line-up of only Hispanic males, none of whom matched the descriptions she provided of the second individual. The same photo of Mr. Flores from the line – up was then repeatedly used in media coverage and described as law enforcement’s suspect in the case. During the trial, Ms. Barganier identified Mr. Flores as the second individual, claiming to be “100% sure” that he was one of the men she saw that day. Ms. Barganier’s identification is the only piece of evidence that places Mr. Flores at the crime scene.
Mr. Childs, whom police knew was involved in Mrs. Black’s murder because of the distinct VW he owned, confessed to shooting Mrs. Black, pled guilty and was sentence to a term of 35 years with parole eligibility after 17 years. Mr. Childs was released in April 2016, around the same time Mr. Flores’ first execution date was scheduled. At the time of trial, Mr. Childs’ plea deal had not been disclosed to defense counsel, or the jury. Under Texas’ law of parties, a person can be charged with capital murder and sentenced to death if they participated in a felony that resulted in a murder, even if they did not kill anyone themselves. The law holds accomplices equally responsible as the person who committed the killing. Mr. Flores was convicted under this statute as a knowing participant in the robbery and murder, but he denies being an accomplice or present, and has only been placed at the crime by Ms. Barganier’s mid-trial identification.
In 2016, Mr. Flores came within five days of execution before his date was stayed by the TCCA to allow him to litigate a claim that prosecutors unconstitutionally obtained his conviction through the use of unreliable “hypnotically refreshed” testimony. According to an affidavit Mr. Flores submitted to the court from Professor of Psychology Steven Lynn, research has linked “hypnotic refreshment” with the creation of false memories. “Clearly, the techniques that were used to refresh [the witness’] memory would be eschewed today by anyone at all familiar with extant research on hypnosis and memory,” Prof. Lynn wrote. Despite this evidence, the state court rejected Mr. Flores’ appeal in 2018, and in 2020, the TCCA upheld the ruling. In 2023, lawmakers passed legislation ultimately banning the use of hypnosis-based testimony in criminal proceedings, as concerns grew over the use of the practice. This law, however, does not apply retroactively and did not provide relief for Mr. Flores, despite lawmakers’ acknowledgement of the unreliability of the technique as evidence.
In June 2025, counsel for Mr. Flores filed a motion arguing he should be granted a new trial because of changes in the understanding of witness testimony and changes to state law regarding the use of hypnotized witnesses since 1999. Among the filings were new expert declarations that cast even further doubt on the integrity of Mr. Flores’ conviction. This petition was dismissed in October 2025 on procedural grounds without addressing the merits of his claims.
The U.S. Supreme Court is expected to consider Mr. Flores’ petition during a conference in May 2026.
