The Journal of Psychiatry & Law 38/Spring-Summer 2010

From false confession to wrongful conviction: Seven psychological processes

BY RICHARD A. LEO, PH.D., J.D. AND DEBORAH DAVIS, PH.D.

A steadily increasing tide of literature has documented the existence and causes of false confession as well as the link between false confession and wrongful conviction of the innocent. This literature has primarily addressed three issues: the manner in which false confessions are generated by police interrogation, individual differences in susceptibility to interrogative influence, and the role false confessions have played in documented wrongful convictions of the innocent. Although the specific mechanisms through which interrogation tactics can induce false confessions, and through which they can exert enhanced influence on vulnerable individuals have been widely addressed in this literature, the processes through which false confessions, once obtained by police, may lead to wrongful conviction have remained largely unaddressed. This article addresses this gap in the literature, examining seven psychological processes linking false confession to wrongful conviction and failures of post-conviction relief: (1) powerful biasing effects of the confession itself, including incorporated misleading specialized knowledge (inside crimerelevant knowledge displayed by the suspect in the false confession, but acquired through outside sources (such as the interrogator)
AUTHORS NOTE: For additional information about this article contact:

Richard A. Leo, School of Law, University of San Francisco, 2130 Fulton Street, San Francisco, CA 94117. Ph: 415-422-6513. E-mail: rleo@usfca.edu.
2010 by Federal Legal Publications, Inc.

10

FALSE CONFESSION

rather than in the course of the commission of the crime); (2) tunnel vision and confirmation biases, (3) motivational biases, (4) emotional influences on thinking and behavior; (5) institutional influences on evidence production and decision making; and inadequate context for evaluation of claims of innocence, including (6) inadequate or incorrect relevant knowledge, and (7) progressively constricting relevant evidence. We discuss reciprocal influences of these mechanisms and their biasing impact on the perceptions and behaviors of suspects, investigators, prosecution and defense attorneys, juries, and trial and appellate judges. KEY WORDS: False confession, wrongful conviction, interrogation.

All evidence to the contrary . . . On July 10, 1997, Danial Williams was arrested and charged with the brutal rape and murder of his neighbor Billy Boskos wife, Michelle. Two days earlier after debarking from his navy ship the USS Simpson, Billy Bosko had come home to find her lying in a pool of blood on their bedroom floor, naked from the waist down and dead from strangulation and multiple stab wounds to her chest (Wells & Leo, 2008). Police soon embarked on a remarkably flawed investigation of the case that eventually led to the wrongful accusation and incarceration of seven men, an investigation and prosecution described by one writer as the present-day equivalent of such insanities as the Salem witch trials of the 1690s (Connery, 2008; see Wells & Leo, 2008 for a full account of the case). Detectives elicited false confessions from four menDanial Williams, Joseph Dick, Eric Wilson and Derek Ticewho would later come to be known as the Norfolk Four. Three of these men (Williams, Dick and Tice) remain incarcerated to this day serving life sentences, despite DNA exculpations, the identification of the true perpetrator

11

(Omar Ballard), his corroborated confession, and the match of his DNA to the semen found in the victim. The fourth, Eric Wilsonwho, unlike the others, had falsely confessed only to rape instead of both rape and murderserved more than seven years before being released from prison in 2005. Three other additional innocents (Richard Pauley, Geoffrey Farris and John Danser) were also arrestedbut did not confess. Each served 7 to 10 months in jail before capital murder charges against them were dismissed. In this article, we begin by briefly recounting how and why police detectives initially targeted the innocent suspects in the Norfolk Four case and then elicited four false confessions. We then identify seven psychological processes that are often involved when false confessions lead to wrongful conviction, using the case of the Norfolk Four to illustrate the way in which these psychological processes may affect the post-confession thinking and actions of those involved in producing a wrongful convictionfrom defendants, investigators and attorneys, through trial judges and juries, to the rulings of appellate courts. The Norfolk Four: From crime to false confession When Billy Bosko found his wife Michelle dead at mid-afternoon on July 8, 1997, he ran to his neighbor Danial Williams house to borrow a phone to call 911. Williams called, and then he, his wife Nicole and his visiting parents went back with Billy to his apartment. Williams helped Billy cover his wifes body with a blanket, and then waited with him for police to arrive. As police began to question neighbors about what they might have witnessed and who might have committed the crime, they encountered Michelles friend, Tamika Taylor. Asked who may have had any reason to kill Michelle, Taylor pointed to Williams, saying that she thought he seemed

12

FALSE CONFESSION

obsessed with Michelle in a creepy and sexual way (Wells & Leo, 2008). Ironically, Taylor also suggested that police investigate her friend Omar Ballard, the true perpetrator. They and others had all partied together at Michelles apartment only four nights earlier, and Taylor said that Williams had danced in a sexually suggestive manner, while allegedly leering at Michelle. This was sufficient for Detective Maureen Evans to conclude that Danial Williams was the person who raped and killed Michelle Bosko. Omar Ballard was never investigated until he confessed to the crime in a letter provided to police by Tamika Taylors mother on February 22, 1999almost 19 months after the murder. Had he been investigated, police would have discovered a very troubling and suspicious history. Only two weeks before Michelle Boskos murder, Ballard had assaulted a girl named Melissa Morse, who lived in the same block. An arrest warrant was issued against Ballard for this assault on the same evening that Danial Williams was interrogated. Ballard had proceeded to rape a 14-year-old girl in the neighborhood only 10 days after Michelles murder, and had been identified and arrested for that crime a week later. He was known to stay with Tamika Taylor, and Taylor had reported to police that Ballard had become friends with the Boskos, that he spent time with Michelle when Billy was at sea, and that he had partied with her, Danial and the others at Michelles place a few days before the murder. Yet police never pursued any investigation of Ballard as a suspect in Michelles murder. In contrast, Danial Williamswho had no police record would seem to be an unlikely suspect. He had been married for only ten days, to a wife who was suffering from cancer, and his parents had been visiting for the last several days. His wife told police that he had been sleeping with her at the time of the murder. Nevertheless, initially police focused only on Danial Williams. He became the primary suspect within the very first hour of the investigation, based solely on Taylors

13

statement about his alleged sexual interest in Michelle. Detective Evans asked him to come to the police station to talk further with her and Detective Halverson about the case. Never imagining that he was a suspect, Williams willingly followed them to the police station, arriving at 6:30 p.m. Police kept him waiting while they followed up with other witnesses until about 8:00 p.m., when they undertook an interrogation that would last through the night, until after 7:00 a.m. the next morning. Williams, who was exhausted and had not eaten since breakfast at around 9:00 a.m., maintained his innocence for the first 10 hours as he was interrogated by three different detectives. He agreed to take a polygraph, and he was falsely told that he had failed.1 He was relentlessly accused of having sexual interest in Michelle and of raping and murdering her; he was accused of lying about his reported memories of his relationship with Michelle and his activities at the time of the murder; and he was told, falsely, that an eyewitness had seen him leave Michelles apartment around the time of the murder. As the interrogation continued into the wee hours of the morning, Williams became increasingly exhausted and sleep deprived; at one point when he put his head down on the table to rest, the interrogators told him to pick it back up. As Williams continued to deny involvement, the interrogators began to suggest that he could have repressed his memories of the crimethat he could have blacked out, or been sleepwalking, that he could have amnesia. 2 Even though he had begun to doubt his own memory and to wonder if he had somehow committed the crime, he continued to resist their repeated accusations. But at around 4:50 a.m., a new interrogator was brought inDetective Robert Glenn Ford. Ford was known for his aggressive interrogation techniques, and his frequent success in inducing confessions. But he was also known for some more shady techniques, many known to enhance the risk of false confession, and had already been demoted once for coercing false confessions from three

14

FALSE CONFESSION

teenagers in 1990. Subsequently, he induced known false confessions in at least two more cases in 1994 and 1997 (Wells & Leo, 2008). Danial Williams reported that during the next hour Ford yelled at him, poked him in the chest, got in his face and told him he knew Williams had murdered Michelle, andas occurred in the interrogations of all the subsequent suspectsthreatened Williams with the death penalty if he refused to confess. Ford stressed that Williams was facing capital murder charges, but that Ford could help him get a lesser charge if he confessed. At about 5:50 a.m., after almost 12 hours in the police station, Williams gave in, and began to concoct a story of his involvement. He later reported that he was exhausted, confused, not feeling well, and finally just couldnt take the pressure anymore as he came to believe that he might have actually committed the crimeso he told the detectives what they wanted to hear because he just wanted the questioning to end. He made up a story that was inconsistent in many respects with the evidence, in which he admitted to things that the detectives acknowledged they had made up. He was influenced to change his story to better fit the evidence to some degree over the next hour, and at 7:00 a.m. the detectives had Williams give an audiotaped confession. This would be the first of many coerced false confessions that Detective Ford and his colleagues elicited over the next couple years. By 11:00 a.m., having found that the autopsy report was inconsistent with many of the details in Williams taped confession, Detective Evans brought him back into the interrogation room to confront him about the glaring major inaccuracies and educate him about the correct details, and soon elicited a changed story and second, revised, taped confession. This sequencecoerced confession, discovery that it couldnt be true, coerced revisionsoccurred repeatedly as each coerced story proved false and police targeted more and

15

more innocent young men, a total of 12 before the case was over. In December of 1997, the DNA analyses came back with a shocker for police. Danial Williams was excluded as Michelle Boskos rapist. Assuming that Williams roommate Joseph Dick must have somehow been involved with Williams in the crime as a coperpetrator, Detective Ford and his colleague Detective Wray hauled Dick in for questioning andafter 9 hours of accusatory interrogation that involved lies about evidence, including false reports of a failed polygraph, and death penalty threatscoerced a false confession from him as well. Police arrested Dick on January 15, 1998, only to again discover in March that Dicks semen did not match that of the perpetrator. Still convinced that Williams and Dick must be guilty despite the absence of physical evidencebecause, they believed, innocent people do not confesspolice now concentrated on locating another co-perpetrator who would match the DNA. They arranged to have an informer placed in Joe Dicks prison cell, who succeeded in encouraging Dick to act in ways that led to another suspectEric Wilson, another sailor and acquaintance of Williams and Dick. Hauling Wilson in for questioning, Detectives Ford and Steve Hoggard used many of the same interrogation tactics as beforeincluding repeatedly yelling at Wilson, accusing him of the rape, attacking his denials, lying to him about false polygraph results, causing him to question his lack of memory of committing the crime and showing him crime scene photos while feeding him details of the crime over the course of a 10-hour overnight interrogation. Like Williams and Dick, Wilson falsely confessed to participating in the rape of Michelle Bosko; however, unlike both of them, Wilson did not confess to the murder because for some reason the detectives had not accused him of it (Wilson later testified that he would have falsely confessed to anything, including the murder, to get away from Detective Ford).

16

FALSE CONFESSION

Wilsons DNA returned as no match to the perpetrator of the rape in June of 1998, almost a year after the murder, nor had his fingerprints been found anywhere in the apartment. Meanwhile, since the preceding two false confessions had made no mention of Wilson, police had reinterrogated Dick, and elicited a new story implicating Wilson. But with the new DNA results in hand, police soon returned to Dick, whose attorney had presumed his guilt all along and had instructed Dick to cooperate in every way with police in order to avoid the death penalty. Detectives accused Dick of still not telling the truth, as they interrogated him to elicit the name or names of the additional coperpetrator(s) they assumed must be involved. Dick now told the story that there were three additional perpetrators involved, but his description and police artist sketch only led to oneDerek Tice. Though Joe Dick had given police the name George, Derek Tice became a suspect because police could not locate any acquaintance of Williams or Wilson named George. Police had questioned Melissa Wharton, an acquaintance of Danial Williams, who had mentioned the name Derek Tice. After locating a photo of Tice and comparing it to the police sketch of George, the detectives concluded Tice was a match, and soon showed Joe Dick a series of photos including Tice. Under pressure, Dick told detectives that Tice was George. Detectives Ford and Wray soon subjected Tice to a 9-hour interrogation, once again using many of the same interrogation tactics as with the three othersrepeatedly accusing him of committing the crime, repeatedly calling him a liar, telling him that he had failed a polygraph, telling him that other suspects identified him as participating and would testify against him, repeatedly threatening him with the death penalty and telling him the only way he could save his life was by confessing, and showing him crime scene photos and feeding him details of the crime.

17

Tice became the final false confessor of the case in mid-June of 1998. This time Ford did not wait to find the results of the DNA analyses to pressure Tice to finger yet more suspects. According to Tice, Ford told him that he knew others were involved, and that if Tice would give him their names, he would see to it that Tice didnt get the death penalty. Tice offered the names of the first two men he could think of who were acquainted with the other defendantsRichard Pauley and Geoffrey Farriswho were promptly interrogated and arrested. But in late August, Tice, Pauley and Farris were all eliminated as rapists by DNA analysis, and neither Pauley nor Farris had confessedthough Joe Dick had now offered a new story implicating both. In September, in the context of negotiating a plea agreement to save him from the death penalty, Tice offered Detective Ford another name, John Danser, as allegedly the seventh person involved in the crime. Although the previous six had now all been exculpated by DNA analysis, police and prosecutors continued to insist they were guilty, and Detective Ford investigated six additional others with no success. Although it soon became clear that Danser had an airtight alibi, and although in February of 1999 DNA results excluded him as the rapist, he was nevertheless prosecuted. In January, 1999, in response to prosecutors demands as part of a plea agreement to avoid the death penalty, Danial Williams had provided a new statement now implicating Dick, Tice, Wilson, Pauley, Farris and Danser (but not Omar Ballard)even though none were mentioned in his original confession. Seven innocent young men were now in jail for the rape and murder of Michelle Boskowith none of their DNA matching that of the semen found in her body. In February of 1999, Ballards letter to Tamika Taylor found its way to police. When interrogated, he admitted the crime within 20 minutes of questioning and maintained that he had acted alone. His statement contained accurate details of the crime not mentioned in those of any of the other defendants.

18

FALSE CONFESSION

But police and prosecutors did not believe he acted alone despite the fact that only he had a violent criminal record, only his DNA had been deposited in Michelle Bosko, and only he could provide accurate crime detailsbut instead believed, absurdly, that they had finally located the missing member of an eight-man gang-rape murder. They offered Ballard a plea contingent on his implicating the others, which he accepted to avoid the death penalty. To maintain this agreement, he provided a new confession also implicating the others, which he later retracted. But police and prosecutors never chose to believe him, preferring instead to believe the far-fetched scenario that, against all physical evidence and logic, he somehow acted in concert with the other seven defendants to brutally rape and murder Michelle Bosko. Danial Williams and Joe Dick pled guilty to avoid receiving the death penalty, which their lawyers had convinced them was almost certain if they went to trial. Eric Wilson and Derek Tice were found guilty at their separate jury trials, and, after his appeal of the first verdict was granted, Tice was found guilty again at a second jury trial. Although Wilson was released after finishing his sentence for rape, Williams, Dick and Tice have been incarcerated for more than 11 years for a crime they did not commit. Why were these multiple false confessions viewed as so compelling and so impossible to discount? Why did police, prosecutors and defense attorneys fail to believe the Norfolk Fours retractions, despite the DNA exculpations, lack of physical evidence, the presence of alibis, the evidence of police coercion, the mountain of inconsistencies, inaccuracies and contradictions in their accounts, their failures to implicate others in their initial confessions (despite the obvious human tendency to distribute blame), the apprehension and confession of the true perpetrator, and many other facts inconsistent with guilt? In the next sections, we identify seven psychological processes often involved in cases where false confessions lead to wrongful conviction

19

and where innocent defendants fail to receive post-conviction relief, and we note their role in the case of the Norfolk Four. Seven psychological processes I. Confession and the role of misleading specialized knowledge Confessions are universally viewed as extraordinarily persuasive evidence of guilt, particularly when they contain a plausible story line, a description of motives, explanations, crime knowledge, emotional expressions, and acknowledgments of voluntariness (Leo, 2008). Former U.S. Supreme Court Justice William Brennans observation that no other class of evidence is so profoundly prejudicial (Colorado v. Connelly, 1986, p. 182) is amply supported by social science research (Miller & Boster, 1977; Kassin & Sukel, 1997; Kassin & Neumann, 1997; Leo & Ofshe, 1998; Drizin & Leo, 2004). Confessions strongly bias the perceptions and decision making of criminal justice officials and jurors alike because most people assume that a confessionespecially a detailed oneis, by its very nature, true. Confession evidence therefore tends to define the case against a defendant, usually overriding any contradictory information or evidence of innocence (Leo & Ofshe, 1998). If introduced against a defendant at trial, false confessions are highly likely to lead to wrongful convictionseven when they are elicited by questionable interrogation methods and are not supported by other case evidence. False confessions have contributed not only to erroneous jury verdicts, but also to the entry of false guilty pleas (Redlich, in press), as occurred in the Norfolk Four case. Because self-incriminating statements are viewed as such inherently compelling evidence of guilt, false confessions are a leading cause of wrongful conviction. In aggregated case studies, they have accounted for 14 to 60% of documented wrongful convictions (Warden, 2003; Bedau & Radelet, 1987; Scheck, Neufeld, & Dwyer, 2000; Gross, Jacoby, Matheson, Montgomery, & Patel, 2005; Garrett, 2008). A confession sets in motion a seemingly irrefutable presumption of guilt among justice officials, the media, the

20

FALSE CONFESSION

public, and jurors (Leo & Ofshe, 1998). This chain of events, in effect, leads each part of the system to be stacked against the confessor; he will be treated more harshly at every stage of the investigative and trial process (Leo, 1996). He is significantly more likely to be incarcerated prior to trial, charged, pressured to plead guilty, and convicted. Moreover, the presence of a confession creates its own set of confirmatory and cross-contaminating biases (Findley & Scott, 2006), leading both officials and jurors to interpret all other case information in the worst possible light for the defendant. For example, a weak and ambiguous eyewitness identification that might have been quickly dismissed in the absence of a confession will instead be treated as corroboration of the confessions validity (Castelle & Loftus, 2001). As the case against a false confessor moves from one stage to the next in the criminal justice system, it gathers more force and the error becomes increasingly difficult to reverse. This process typically starts with the police. Once they obtain a confession, they typically close their investigation, deem the case solved, and make no effort to pursue any exculpatory evidence or other possible leadseven if the confession is internally inconsistent, contradicted by external evidence, or the result of coercive interrogation (Leo, 1996; Ofshe & Leo, 1997a, 1997b)for once they elicit a confession it serves to confirm their presumption of guilt. In order to maintain their presumption of guilt, police will disregard or misinterpret evidence of innocence. Even if other case evidence emerges suggesting or even demonstrating that the confession is false, police almost always continue to believe in the suspects guilt and the accuracy of the confession (Leo & Ofshe, 1998; Drizin & Leo, 2004). All of this occurred in the Norfolk Four case. Once police had elicited a confession from Danial Williams, they refused to acknowledge the possibility of his innocence, despite considerable evidence demonstrating it. In Williams first

21

confession, several key details were wrong. For example, he said he hit Michelle Bosko with a hard-soled shoe that he picked off the floor, though Michelle had suffered no injuries to her head; he did not recall seeing any blood, though there was, in fact, plenty at the scene; he said that Michelle was still alive and screaming when he left the apartment, but this would have been impossible. Williams said nothing about stabbing Michelle because clearly he did not know that was how she had died. After Williams initial taped statement, Detective Evans went to the autopsy, where she learned of the errors in Williams confession. She then reinterrogated Williams and literally educated him about the correct details of the crime, telling him, for example, that Michelle had died of stab wounds, that she had been stabbed in the chest and that he needed to provide a more accurate statement, which he subsequently did. Rather than considering the possibility that Williams might be innocent, Detective Evans became even more certain of Williams guilt because Innocent people dont confess to crimes they didnt commit (Wells & Leo, 2008, p. 36). Still, Williams second taped statement contained additional factual errors, and Williams still could not describe the knife used to kill Michelle Bosko. This pattern repeated itself throughout the Norfolk Four case. When Williams DNA did not match the sperm, blood or other genetic material recovered from the crime scene, police only dug in deeper and assumed he must have raped Michelle Bosko with an accomplice who had ejaculated inside her. When police coerced a confession from Joseph Dick after another lengthy interrogation, they never entertained the possibility that he may be innocent, despite the fact that his account was also inconsistent with and contradicted by key crime facts. When Dicks DNA too did not match the sperm, blood or other genetic material recovered from the crime scene, police again assumed that Dick (and Williams) had been involved with yet another accomplice who had ejaculated inside Michelle. The same pattern repeated itself after they elicited confessions from Eric Wilson and Derek

22

FALSE CONFESSION

Tice (again after lengthy and coercive interrogations), and they also never entertained the possibility that Wilson or Tice may be innocent despite the fact that their confessions too contained obvious errors that contradicted the crime facts, and their DNA also did not match the ejaculate left inside Michelle or any of the genetic material described at the crime scene. Moreover, that each confessors statement failed to mention the subsequent confessorWilliams confessions had not mentioned Dick, Wilson or Tice; Dicks had not mentioned Wilsons or Tice; and Wilsons had not mentioned Ticealso did not shake the detectives tenacious belief that each factually flawed confession that they had elicited must somehow be true. Finally, when Omar Ballard confessed that he alone committed the crimewhich was supported by the DNA evidence, since only a single DNA profile was found inside Michelle and it was Ballardspolice could not accept his initial account that Williams, Dick, Wilson and Tice were innocent of the rape and murder. Instead, even though they knew the crime scene was inconsistent with a gang attack, they coerced Ballard into placing the other defendants at the scene with him as a condition of receiving a plea bargain that would save him from the death penalty. The presumption of guilt and the tendency to treat more harshly those who confess extend to prosecutors. Like police, prosecutors rarely consider the possibility that an innocent suspect has falsely confessed. Some are so skeptical of the idea of police-induced false confessions that they stubbornly refuse to admit that one occurred even after DNA evidence has unequivocally established the defendants innocence (Kassin & Gudjonsson, 2004). Once a suspect has confessed, prosecutors tend to charge him with the highest number and types of offenses (Cassell & Hayman, 1996), set his bail higher (Leo & Ofshe, 1998) (especially in serious or highprofile cases), and are far less likely to initiate or accept a plea bargain to a reduced charge (Leo & Ofshe, 1998). The confession becomes the centerpiece of the prosecutions case, and they aggressively prosecute the defendant. When they

23

have confession evidence, prosecutors rarely back down, even in the face of considerable exculpatory evidence. Like the police detectives in the Norfolk Four case, the prosecutors presumed each confessor must be guilty simply because they had confessedinnocent people do not confess prosecutor Valerie Bowen would tell the jury in Eric Wilsons trialdespite the factual errors and inconsistencies in each of the confessions, the DNA and physical evidence that did not support the confessions, and the logic and factual evidence that eventually established Ballard as the sole perpetrator of the rape and murder of Michelle Bosko. Instead, like the police detectives, prosecutors repeatedly disregarded and misinterpreted evidence, including rock solid alibi evidence of innocence. They also repeatedly failed to turn over exculpatory evidence to defense counsel, who accused them of making blatantly false representations on the record to support the prosecution of the innocent men. Like the police detectives in this case, prosecutors attempted to rationalize away obvious contradictions and inconsistencies. For example, when Ballard indicated in his initial confession that he had acted alone, prosecutor D.J. Hansen stated that Ballard, who belonged to a Black prison organization, was lying because he didnt want to share the crime or the credit for a rape-murder with white guys. Prosecutor Valerie Bowen would claim that there was a lot of corroborating evidence in support of the defendants confessions. I cant even go into all of [it] now, it would just take too long, she said. And D.J. Hansen asserted, everything fits together . . . It just all makes so much sense when you look back on it (Wells & Leo, 2008, p. 184). Even defense attorneys tend to presume confessors are guilty and treat them more harshly. As the California Supreme Court has noted, the confession operates as a kind of evidentiary bombshell which shatters the defense (People v. Cahill, 1993, p. 497). As a result, defense attorneys often pressure confessors to accept a guilty plea to a lesser charge

24

FALSE CONFESSION

in order to avoid the higher sentence that will inevitably follow from a jury conviction (Nardulli, Eisenstein & Fleming, 1988). In the Norfolk 4 case, the attorneys for Danial WilliamsDanny Shipley and Robert Frank assumed he must be guilty because he had confessed and took to calling him Denialto his face, to his family and one time even at a court hearingevery time he asserted his innocence. At Williams suppression hearing, Shipley and Frank never raised the possibility that Williams confession was false nor did they point out any of the blatant discrepancies between it and the physical evidence. Joseph Dicks attorney, Michael Fasanaro, accused his client of lying when he told him he was innocent, and told Dicks parents, without a doubt, Joe was involved in every aspect of the case (Wells & Leo, 2008, p. 84). This persuaded Dicks family members that he was guilty, and as a result they also pressured Dick to accept responsibility for the crime, enter into a plea agreement and testify against the other defendants in order to avoid the death penalty (Williams attorneys had also persuaded his family that he too was guilty). Fasanaro, who never investigated Dicks provable alibi that he was out to sea at the time the murder had occurred, dismissed the inconsistencies and blatant errors in Dicks many confessions as meaningless and predictable for a criminal defendant. Alan Zaleski, Tices first lawyer, also assumed his client was guilty because he couldnt understand why Tice would confess if he was not (Wells & Leo, 2008). American judges also tend to presume that confessors are guilty and treat them more punitively. Conditioned to disbelieve defendants claims of innocence or police misconduct, judges rarely suppress confessions, even highly questionable ones (Givelber, 2000). Thus, if the case does go to trial, the jury is highly likely to learn of the confession, as they did in the trials of Tice and Wilson. Even in the absence of other inculpatory evidence, and even if elicited through a clearly coercive interrogation, as was the case for the Norfolk

25

Four, false confessions are highly likely to lead to wrongful convictions if introduced against a defendant at trial, as they did for Wilson and Tice. In their study of 60 false confessions, Leo and Ofshe (1998, 2001) found that 73% of the false confessors whose cases went to trial were erroneously convicted; 81% were in Drizin and Leos (2004) study of 125 false confessions. This rate of wrongful conviction becomes even larger when incorporating the number of false confessors who plead guilty rather than take their cases to trial (78 and 85%, for the two studies, respectively). Despite their demonstrably false confessions, both Williams and Dick pled guilty to the rape and murder of Michelle Bosko to avoid the death penalty. The findings from the studies of aggregated false confessions cases are consistent with those from experiments and public opinion surveys (Leo, 2008). They all point to the same conclusion: that a confession is uniquely potent (Kassin & Neumann, 1997, p. 469) in its ability to bias the trier of fact in favor of the prosecution and lead to a wrongful conviction (Leo and Ofshe, 1998). Experimenters have demonstrated that mock jurors also find confession evidence more incriminating than any other type of evidence (Kassin & Neumann, 1997; Miller & Boster, 1977). Jurors fail to appropriately discount false confession evidence, even when the defendants confession was elicited by coercive methods and the other case evidence strongly supports his innocence. In one mock jury study, Kassin and Sukel (1997) found that confessions greatly increased the conviction rate even when mock jurors viewed them as coerced. Most Americans simply accept confession evidence at face value. Randall McFarlane, the jury foreman in Derek Tices first trial, subsequently explained that Tices confession just washed everything else away. . . . That was the supernova circumstance of the entire trial. It overwhelmed everything else (Wells & Leo, 2008, p. 228). When false confessors subsequently retract their confessions, they are often not believed, or their retractions are perceived as further evidence of their deceptiveness and

26

FALSE CONFESSION

thus guilt (Ofshe & Leo, 1997a)as occurred in the Norfolk Four case each time Williams, Dick, Wilson, or Tice attempted to retract their confessions. The weight of the false confession continues to sink the confessors hopes of exoneration or less severe outcomes at all points subsequent to conviction. This begins at sentencing, as trial judges tend to punish defendants with harsher sentences for their claims of innocence, and the consequent costs to the state in time, effort and resourcesand for failing to express remorse or to apologize (Leo, 2008). False confessions likewise compromise the defendants efforts at post-conviction relief. It is exceedingly rare that criminal justice officials will take seriously claims of false confession and wrongful conviction (Leo, 2008; Bedau & Radelet, 1987). The system provides no regular mechanisms for reviewing the substantive basis of convictions, officially presumes the defendants guilt after he is convicted, treats the jurys verdict with deference, and interprets any new evidence in the light most favorable to the prosecution rendering the correction of any mistaken conviction unlikely (Leo, 2008; Gudjonsson, 2003; Medwed, 2004). Despite the use of DNA to exonerate innocent prisoners in the last two decades and the increasing number of documented proven wrongful convictions (with and without DNA), criminal justice officials and courts still tend to presume the validity of confession-based convictions. To this day, police and prosecutors in the Norfolk Four case continue to insist that Williams, Dick, Wilson, and Tice (as well as Farris, Pauley and Danser) all raped and murdered Michelle Bosko, despite the overwhelming mountain of physical and logical evidence to the contrary.
The role of misleading specialized knowledge

The use of misleading specialized knowledge (MSK) occurs when police investigators feed the suspect unique nonpublic crime factsfacts that are not likely guessed by chanceand then insist that these facts originated with the suspect

27

(Gudjonsson, 2003; Leo, 2008). Sometimes referred to as guilty or inside knowledge, such facts may include the nature of the murder weapon, the location of wounds, placement of objects within the crime scene, and other unique nonpublic details of the crime (see Leo, Drizin, Neufeld, Hall, & Vatner, 2006). When included in the suspects postadmission narrative, the facts are believed to reveal that he possesses information that only the true perpetrator would know and therefore he must be guilty. The inclusion of such guilty or inside knowledge would seemingly be possible only if the suspect had participated in or fully committed the crime, and they therefore lead observers to conclude he must be guilty. In fact, such MSK details are typically incorporated into the statements of innocent defendants because during the interrogation(s) police mention them, show them crime scene photos, or actively argue with the suspect that the crime occurred in a particular way (Leo, 2008). While true guilty knowledge can be a useful aid for conviction of the guilty, MSK is a truly pernicious influence, in that it so effectively implicates the innocent in the eyes of those who investigate and judge him. Police rely on MSK to convince prosecutors of the validity of the confession, who in turn rely on it to convince jurors of the defendants guilt. Judicial rulings on the voluntariness and admissibility of the confession can be prejudiced by judges views of its validity, as judges are loath to suppress a confession that may be the only means of conviction of a person they believe to be clearly guilty. Defense attorneys are influenced by MSK to believe their clients are guilty and then use it to pressure their clients to plead. And finally, appellate courts may rely in part on MSK in support of decisions to uphold trial convictions (Leo, 2008). The dangers of MSK are particularly severe when the interrogation is unrecorded. The defense is in the difficult position of having to show that all MSK was first mentioned, and fed to the suspect, by the policein other words, that the

28

FALSE CONFESSION

defendant learned all the guilty knowledge from the police, rather than in the course of committing the crime. Absent a recording, the source of the information can only be evaluated in light of the accounts given by police and the suspect. Even with the best of intentions, memories of both are likely subject to failures of memory for the contentand crucially, the orderof what was said in interrogations lasting from one hour to more than a full day (see Davis & Friedman, 2007, on memory for conversation). But these intentions can be overridden by self-protective motivations arising as a result of challenges to the validity and voluntariness of the confession. In many documented cases of false confession leading to wrongful conviction, police have refused to acknowledge that they fed the false confessor the MSK used to make his confession so credibleeven when the confession had incorporated facts that had been believed by police during the interrogation, but later been proven false (Leo, 2008). Motivational biases among police may combine with genuine failures of memory to obscure the true source of MSK, even as early as during the interrogation, and as late as after the suspect has definitively been proven innocent. In the case of the Norfolk Four, the detectives failed to record their lengthy interrogations of Williams, Dick, Wilson and Tice, but recorded a brief confession statement at the very end of each. The defendants alleged that detectives educated them about the crime scene facts during each of these interrogations. Each of the Norfolk Four incorporated multiple instances of MSK into their confessions. Ford and the other detectives had shown them crime scene photos, offered their own theories of the case, told the suspects what they already knew, and coerced them into repeatedly changing their accounts to fit the detectives current understandings of the facts. The result was that the confessions included details thatabsent contamination by the detectives during the unrecorded interrogationswould ordinarily seem to corroborate inside knowledge of a crime. The fact that the defendant had given detailed confessions,

29

however, was taken as further evidence of their guilt. In addition to being fed facts, however, some of the defendants had also learned of details of the crime prior to their interrogation. For example, by the time Tice was interrogated, he already knew details he had acquired from listening to news accounts, attending Williams preliminary hearing, and visiting Williams in jail. Unfortunately, neither the detectives nor prosecutors realized the source of the four suspects knowledge of the crime and, instead, attributed it, misleadingly, to inside guilty knowledge. Because there were no recordings to show the process through which detectives had contaminated the suspects with these facts, the prosecutors and juries mistakenly interpreted the details in the confessions to corroborate their veracity. II. Tunnel vision and confirmation biases Tunnel vision refers to the tendency to selectively focus attention on one target to the exclusion of all others. In the criminal justice system, tunnel vision has been used to refer to the tendency to focus attention on one suspect (or group of suspected coperpetrators), and on evidence relevant to his (or their) guilt. We suggest that the problem of tunnel vision extends farther within the legal system, such as, for example, when a defense attorney may become focused on one legal goal (for example, preventing a clients execution), ignoring evidence and lines of defense relevant to the goal of exoneration. Though cognitive researchers are apt to define tunnel vision strictly in terms of the focus of attention (e.g., Safer & Christianson, 1993), legal scholars often include confirmation biases in their notions of tunnel vision. Confirmation biases broadly refer to information processing and behavioral biases leading a person to selectively attend to, seek out, produce, and interpret evidence in ways that support existing beliefs, perceptions, and expectations, and to avoid or reject evidence that does not (Gilovich, 1991). This includes self-fulfilling expectations, whereby the person engages in behaviors causing his or her expectations to be (or appear to be)

30

FALSE CONFESSION

fulfilledfor example, when hostile behavior toward a suspect leads the suspect to act anxious and evasive, thus appearing to confirm expectations that he is guilty and trying to hide it. Hence, in a more inclusive view of tunnel vision in the criminal justice system, Martin defines it as the tendency to focus on a suspect, select and filter the evidence that will build a case for conviction, while ignoring or suppressing evidence that points away from guiltthus incorporating confirmation biases as part and parcel of tunnel vision (Martin, 2002, p. 848). The basis of both, of course, is strong beliefs or expectations that first lead to selective focus on a particular individual or class of suspects (tunnel vision) and then promote selective search for, production, and interpretation of evidence (confirmation biases). Tunnel vision and confirmation biases are implicated pervasively in the creation and consequences of false confessions, beginning with their influence on the necessary first step toward wrongful convictionthe misclassification of an innocent person as guilty. Among potential triggers of such misclassifications are beliefs based on gut hunches (e.g., the suspect is displaying suspicious demeanor or emotions) or erroneous assumptions (Ofshe and Leo, 1997a)such as the assumptions that led to the targeting of Danial Williams (i.e., that a person who is sexually attracted to a victim will often kill her if rejected)and crime-related schemas or intuitive profiles of likely perpetrators (e.g., women are typically murdered by their mates; Davis and Follette, 2002). Such assumptions can trigger tunnel vision and an initial assessment of enhanced probability (or a definite classification) of guilt. Such an intuitive theory of the motivation for a rape/murder such as Michelle Boskos first guided Detective Evans to focus on Danial Williams. Later this uncorrected assumption of his guilt led to the exclusive focus on a series of others

31

with only the probable cause that they were acquainted or associated with Williams. They ignored Taylors suggestion that they investigate Ballard, given on the day of the murder discovery. They didnt pick up on Ballards arrest for similar crimes in the same neighborhood. This tunnel vision kept all attention off the true suspect until he was forced upon police by his own confession letter. Once targeted as a suspect, confirmation biases began to affect judgments of Williams behavior almost immediately. For example, Detective Evans interpreted his willingness to come to the station to talk to them and failure to protest as overly cooperative, and therefore suspicious. She also interpreted the bumper sticker on his car Sex is like pizza. When its good, its very good. When its bad, its still pretty good!as evidence of sexual perversion. A second trigger of misclassification can occur during the preinterrogation Behavior Analysis Interview in which the now suspected perpetrator is assessed for deception. Flawed training in behavioral analysis encourages police to use erroneous and misleading criteria for assessment of deception (see Vrij, 2008), yet to mistakenly believe that they detect deception with 85% or better accuracy (Kassin & Fong, 1999; Meissner & Kassin, 2004). But behavioral confirmation biases immediately come into play as the stressful interview and interrogation lead the suspect to display anxious behaviors investigators are erroneously trained to interpret as guilt. During the subsequent interrogation, interrogators assume guilt, seek only confirming statements, and actively discourage and ignore exculpatory denials, verbalizations of innocence, and explanations. As Findley and Scott (2006) point out:
The very notion of an interrogation, therefore, expressly embraces the foundational problems with tunnel visiona premature conclusion of guilt, and an unwillingness to consider alternatives. In this context, however, the tunnel vision is not inadvertent,

32

FALSE CONFESSION

but deliberate; police are taught that this is the way to advance their investigation. Cognitive biases are openly encouraged. (p. 335)

The integral roles of tunnel vision and confirmation biases in promoting false confessions have been repeatedly documented in aggregated case studies (Leo & Ofshe, 1998; Drizin & Leo, 2004), experimental studies (Kassin & Fong, 1999; Kassin, Goldstein & Savitsky, 2003; Meissner & Kassin, 2002), and documentary studies (Ofshe & Leo, 1997a, 1997b) of police interrogation and false confession. Likewise, these biases infiltrate the post-admission narrative portion of the confession, in which the suspect narrates the means of commission and motive underlying the crime. During this stage, with confidence in the suspects guilt cemented by his initial confession, detectives often engage in suggestive and leading questions and challenges to the suspects account that continue to shape the confession to confirm detectives current understanding of the crime, to fit evidence currently known to the detectives and to build an ever more airtight case against the suspect (Leo, 2008). The confession itself then becomes a significant source of continuing tunnel vision and confirmation biases. With belief in guilt more firmly entrenched, investigators treat the confession as confirmation of their belief in the suspects guilt rather than as a hypothesis to be tested against case evidence. Further investigation tends to stop, and the confession becomes the centerpiece of the case as it is conveyed to prosecutors, judges, juries, and in postconviction appeals. The formal presumption of innocence is replaced by an unshakable assumption of guilt that provides the context for interpretation of all other evidence (Leo & Ofshe, 1998). Inconsistent evidence tends to be ignored, dismissed as flawed, or reinterpreted as either irrelevant or supportive of guilt, by investigators, judges, juries, and prosecutors (who sometimes even suppress contradictory or exculpatory evidence, and, even when the suspect is definitively exonerated, may refuse to acknowledge his

33

innocence or admit any mistakes in the investigation or prosecution). Detective Evans was confronted immediately with evidence from the autopsy of Michelle Bosko that was incompatible with Williams confession. Yet, assuming that innocent people dont confess to crimes they didnt commit, she didnt question his guilt, and instead simply assumed that he lied, forgot details, or got confused. As became standard operating procedure when the other defendants confessions didnt fit the facts, she simply returned to confront Williams with the discrepancies and solicited a new account, feeding him facts she had learned from the autopsy and shaping his new account through several initial versions until she got one that fit her new view of what happened. The tunnel vision and confirmation biases of prosecutors stem from many sources: the institutional and political culture of their offices (Medwed, 2004), role pressures and conviction psychology (Fisher, 1988; Findley & Scott, 2006). Even defense attorneys sometimes succumb to tunnel vision and confirmation bias once they learn that their client has confessed, ruling out the possibility of innocence, as clearly illustrated by the case of the Norfolk Four. As noted earlier, juries too allow the power of confession evidence to bias their judgments (Kassin & Sukel, 1997), leading them to presume guilt, selectively focus on guilt supportive evidence, and selectively ignore and discount evidence of innocence in false confession cases (Leo & Ofshe, 1998). In short, the defendant is subjected to a progressing cascade of guilt presumptive tunnel vision and confirmation biases among all who judge him. The impact of these biases becomes ever stronger as the range of contextual information that could otherwise provide the basis for more accurate evaluation of the validity of the confession shrinks. That is, much of the original information relevant to the suspects guilt is lost as it is selectively filtered first by investigators,

34

FALSE CONFESSION

and later selectively reported by attorneys to judges, juries and appellate courtsand the evidence that remains is interpreted without the full relevant context for judging its validity. III. Motivational biases A large body of literature addressing motivated cognition (e.g., Psychological Inquiry, 1999; Special Issue on Motivated Cognition) or goal pursuit (e.g., Hassin, Uleman, & Bargh, 2005; Elliot & Dweck, 2005) has examined the manner in which current goals direct attention, information search, and processing and interpretation of incoming informationas well as goal-relevant behavior. Goals can be regarded as one source of tunnel vision, in that they selectively direct attention to goal relevant targets and information, and, like confirmation biases, promote biased processing of incoming information but with interpretative biases that would serve the perceivers goals, rather than simply confirm his or her expectations. Finally, goals also motivate behavior in their service, such that goal-promoting behavior becomes more likely. Goal-related biases pervade the processes leading to false confession and wrongful conviction. The primary goal of investigators and prosecutors should be accuracyidentifying and convicting the guilty while making sure to avoid prosecution of innocents (Thomas, 2008). Unfortunately, competing personal, institutional, and external sources (police supervisors, prosecutors, victims and their families, politicians, media, and the pressures of high case loads) create pressures to solve crimes quickly and efficiently, especially in serious and high-profile cases (Findley & Scott, 2006). Such pressures in turn may promote a rush to judgment and escalate the risk of misclassifying innocent persons as likely perpetrators. Once a suspect is so classified and subject to interrogation, still in the grip of these pressures, the police interrogators goals are to induce the suspect to provide incriminating statements (preferably a full confession) that will facilitate a certain and efficient conviction (Leo, 2008).

35

The suspects own goalswhether to escape an intolerable interrogation, to establish innocence, or to minimize the consequences of being implicated in the crimedirect attention to, and processing of, incoming information and retrieval of relevant existing knowledge, as well as his behavioral choices in pursuit of one or more of these goals. An explicit goal of interrogators, as recommended and explained in interrogation manuals (e.g., Inbau, Reid, Buckley, & Janyne, 2001) is to control the goals of the suspect during interrogationsuch that he will give up on the goal of establishing innocence and focus on the goal of achieving the least serious consequences of his guilt. Toward this end, interrogators display absolute and unwavering confidence in the suspects guilt, accept no denials or evidence of innocence as valid, and present true and false evidence ostensibly proving the suspects guilt beyond doubt (Leo, 2008). Once the suspect is convinced by these behaviors that establishing his innocence is hopeless, and he begins to focus on the goal of minimizing consequences, the interrogator can more effectively convince the suspect to confesswhich is clearly not a way to establish innocence, but plausibly can be regarded as a way to minimize consequences (particularly in the context of the interrogators claims regarding the need for and benefits of confession). The suspects goal of minimizing consequences leads him to selectively search for and attend to any clues as to how this may be accomplished. The interrogators good will becomes more important because, as the interrogator has strategically implied (see Kassin & McNall, 1991; Ofshe & Leo, 1997a,1997b; Davis, Weaver & Leo, 2008; Davis, Leo & Follette, 2008), he is often viewed as a means to achieve greater leniency. Suspects naturally assume the detective has greater knowledge and experience concerning the likely consequences of confession versus denial, and therefore give weight to his arguments and implications regarding the reasons why confession will result in greater leniency; and they rely on his explicitly stated or

36

FALSE CONFESSION

clearly implied desires to help the suspect achieve the best outcomes (Davis & Leo, 2006; Follette, Davis & Leo, 2007; Ofshe & Leo, 1997a, 1997b). Next in line are the goals of the prosecution and defense attorneys. The goals of the prosecution to secure a conviction are likely to promote guilt-presumptive tunnel vision and confirmation biases, as discussed earlier. Unfortunately, the goals of the defense attorney can likewise reflect the presumption of guilt. Defense attorneys can feel hopeless, and just as the defendant may already have done in the interrogation room, the defense attorney can shift his goals toward minimizing sentencing, rather than establishing innocence. The result is that defense attorneys may selectively focus on pursuing evidence that may minimize a suspects potential sentence while ignoring and failing to investigate evidence that would support a claim of innocence (Wells & Leo, 2008). In turn, jurors are affected by their own goals during trial. While many may strive to achieve the most accurate verdict, jurors can be subject to goals secondary to a presumption of guilt raised by the confession, as well as to emotions (see next section) triggered by the nature of the crime or efforts of the prosecutor to raise and exploit themsuch as, for example, when goals of punishment and revenge are motivated by anger, fear or disgust, and, as a result, evidence is reviewed in a manner supporting and justifying a verdict of guilt or a punitive sentence of death (e.g., Finkel & Parrott, 2006; Jones & Fitness, 2008; Law and Human Behavior (2006) Special Issue: Emotion in Legal Judgment and Decision Making).
Escalating commitment and the roles of self-protection and selfjustification

As the case proceeds, police detectives and prosecutors will have devoted more and more resources to proving the suspects guilt, and made more and more public statements asserting that guilt and attempting to convince others of it. Escalating commitment (e.g., Tavris & Aronson, 2007) refers to the process by which each action taken in support of

37

a position or goal tends to enhance belief in its validity or importance. Evidence contradicting the validity of the belief or goal becomes increasingly more dissonant (Festinger, 1957) and aversive. Evidence contradicting a strongly held position may also threaten the persons sense of competence or self-esteem, and thereby lead a person to engage in strong self-protective and self-justifying motivations (e.g., Tavris & Aronson, 2007). As a result the person may become hostile to the source of such evidence and to dismiss it as invalid. Such self-protective and self-justifying motivations are often reflected in overwhelming resistance to disconfirmation of suspects guilt once an interrogator has induced a false confession and facilitated the suspects prosecution (Tavris & Aronson, 2007). Similar motivations are likely responsible for the great reluctance of prosecutors to acknowledge exculpatory evidence, particularly when it indicates they may have convicted an innocent man (Medwed, 2004). The case of the Norfolk Four was pervaded by refusal to recognize mistakes, even in the face of overwhelming evidence of innocence. Danser, for example, had an unshakable airtight alibi, verified by multiple parties, that he had been in Philadelphia at work and other activities at times that would have made his presence in Norfolk at the time of the murder impossible. But prosecutor Valerie Bowen later falsely claimed that he could have somehow driven to Norfolk in time and that the time sheets and credit card documents verifying his alibi were hearsay and therefore not admissible. When the trial judge asked prosecutor D.J. Hansen at a sidebar, You dont really believe this man was there, do you? Hansen replied, The commonwealth takes the position that he was there. (Wells & Leo, 2008, p. 177). This was impossible. Detectives and prosecutors explained away exculpatory evidence from autopsy reports, alibis, crime scene forensics, incredible confessions, and other sources to maintain their belief in the legitimacy of their years of investigating and prosecuting men who were clearly innocent.

38

FALSE CONFESSION

It was not until outside eyes began to look at the case that many who doubted the defendants innocencefrom family members to jurorsbegan to question the validity of the multiple confessions. These outsiders, of course, had not been involved in the investigations and prosecutions and had no selfjustification motives driving their judgments. After Tices second conviction, the first author (Leo) solicited the involvement of Peter Neufeld and the Innocence Project to help free the Norfolk Four. This led to a chain of events in which the case was publicized in television and news accounts, and in pleas for clemency to the Virginia governor (see Wells and Leo, 2008 for details). It wasnt until they were exposed to these accounts that many family members, jurors, and justice officials began to believe in the Norfolk Fours innocence. Some family members finally began to believe in the innocence of their own, and jurors began to question their earlier verdicts. In fact, in January of 2006, lawyers released a letter from nine jurors from the Wilson trial urging Virginia Governor Mark Warner to grant clemency to the Norfolk Four. The jurors wrote that after being exposed to material that hadnt been available in trial, they now firmly believed in the innocence of all four defendants. Nevertheless, the police and prosecutors responsible for the investigations and prosecutions remained convinced of the defendants guilt and adamantly opposed to clemency. Even some defense attorneys couldnt shake the forces of selfjustification. Joe Dicks attorney, Michael Fasanaro, Jr., who had always assumed him guilty, failed to investigate his alibi (which could have demonstrated that Dick was on duty the night of the murder) and never pursued evidence of innocence. He also refused to admit (or failed to remember) that Dick had ever told him he was innocent, though when asked to examine his notes, they indicated he had. And he continues to assert that Dick is guilty of the murder despite the overwhelming evidence to the contrary. IV. Emotion Many actions and judgments entailed in the elicitation and consequences of a false confession occur in the context of strong emotion. This has been discussed among interrogation

39

scholars largely with respect to the emotional state of the suspect prior to and during the interrogation. The suspect may be experiencing strong emotions such as grief, distress, or fear as a result of the crime itself, particularly when interrogated shortly after the crime, as was Danial Williams in this case. Police interrogators are trained to use emotions to their advantage. The Inbau et al. (2001) manual, for example, suggests the use of anxiety and guilt to promote confession. Interrogation scholars have identified stress induced confession as those in which the suspect has become so distressed (tired, fearful, anxious, or distressed by the aversiveness of the interrogation) that he becomes willing to do or say anythingincluding giving a false confessionto escape the interrogation (Ofshe & Leo, 1997a, 1997b; Davis & ODonohue, 2004). In addition to motivating escape, intense stress also impairs thinking, and can render the person more susceptible to influence. Studies of decision making under stress have shown, for example, that stress can promote a form of tunnel vision, in which attention narrows and the person relies on less and less relevant information when making the decision (see Easterbrook, 1959; Deffenbacher, Bornstein, Penrod, & McGorty, 2004; Hammond, 2000) At first, as arousal increases, attention narrows to exclude irrelevant information and to focus only on the task or decision at hand. This can result in improved performance, when distractions are ignored. But when stress and arousal continue to increase, the further narrowing of attention begins to result in the exclusion of relevant information, and decisions become poorer. Police are trained to direct the suspects attention on themselves, their arguments and evidence, andthrough accusation, confrontation, and controlling the focus of interrogationto prevent suspects from asserting denials, verbalizing innocence or considering alternatives to confessing (such as asking for an attorney) (Ofshe & Leo, 1997a, 1997b; Davis & ODonohue, 2004). When the suspect is compromised by stress, he becomes even less able to think

40

FALSE CONFESSION

beyond the interrogators suggestions, to retrieve relevant information from memory, and to reason with it to come to a more beneficial decision than the detectives suggestion of confession (Follette, Davis, & Leo, 2007). Moreover, the psychology of emotion literature has indicated that specific emotions can promote specific goals, such as when fear promotes the goal of escape from danger (Davis & Loftus, 2009; Levine & Pizzarro, 2004). Thus, the emotionproduced goal then promotes goal relevant thinking and behavioras when the suspects fear of the consequences of being implicated in the crime leads him to pay particular attention to the means of minimizing the consequences. This renders him particularly susceptible to the detectives suggestions about how confession will result in the least serious consequences. But emotions also affect other parties. Exposure to the aftermath of violent crimes can likewise generate strong emotions in the police who investigate them, leading to the potential of narrowing attention to a more restricted range of suspects, and to anger against the suspects who are identified. Anger, associated with goals of punishing or removing the source (Levine & Pizzarro, 2004), can then promote other forms of tunnel vision serving these goals (proving the suspect guilty, getting a confession, and facilitating punishment). Emotions can also affect the reasoning and goals of attorneys. For example, the attorneys defending Danial Williams and Joe Dick had recently lost death penalty cases, and were extremely fearful of losing another client to the death penalty. Their unwavering focus on avoiding the death penalty through pursuit of plea agreements and cooperation with the police, and their failure to pursue claims or investigations of innocence, may have derived in part from this fear. The judges and juries that hear the cases at trial and beyond also experience emotions that can affect their judgments

41

and indeed attorneys from both sides deliberately attempt to raise and exploit such emotions. As a trial consultant for many years, the second author (Davis) always advised clients as to strategies to make the jury want to find in the clients favorfor example, raising sympathy for an injured plaintiff to motivate the jury to award damages (even if the liability issues dont warrant a verdict for the plaintiff). With such motivation, evidence will be interpreted in a more favorable light, and stricter criteria of proof will be invoked for the jury to find for the opponent. Particularly if ones evidentiary case is weak, jury motivation may be the only path to victory. In the case of a heinous violent crime, the prosecutor is likely to attempt to inflame the jurys passions with horrifying accounts and exhibits of the crime, and damning depictions of the motivations, character and actions of the defendant. At minimum, such emotions, once raised, will inspire the jury to require better evidence of innocence before freeing someone who might be such a perpetrator (see Law and Human Behavior, 2006. Special issue: Emotion in Legal Judgment and Decision Making). When raised in the context of a case where the defendant has confessed, such emotions, to the extent they promote tunnel vision and confirmatory biases, will narrow attention and impair recognition of relevant contextual evidence that could have led to a false confession (e.g., Davis & Loftus, 2009). V. Institutional influences on evidenceproduction and decision making Institutional influences affect the actions of police, attorneys and judges, often with the effect of restricting collection of, access to, or reactions to evidence. Earlier, we noted pressures to solve and prosecute crimes quickly and efficiently. For example, restricted financial resources or other financial incentives operating for the various parties can profoundly impact the collection and use of evidence. Finances can restrict funds for the original investigation of the case by police and prosecution, or the quality of attorneys hired to defend the suspect and their investigations. Financial concerns can affect the decisions of judges who must authorize defense expenditures for court-appointed attorneys

42

FALSE CONFESSION

and their wages, investigation, hiring of appropriate experts, performing forensic tests and examinations, and so onor the decisions of family who must pay for a privately funded defense. As with most biases, guilt-presumptive thinking can affect such financial decisions, without the awareness of those making them. Excessive case loads can also affect the attention devoted to a specific case, as well as discourage pursuit of lines of defense viewed as more time consuming or less likely to succeed. Defense attorneys also face unique incentives associated with case-specific payment arrangements. If the attorney receives a flat fee, he or she may be motivated to minimize the effort and expense of the case by avoiding trial if possible, or to maximize profit by minimizing the resources devoted to trial (such as for hiring experts and investigators, ordering tests of evidence, etc.)goals in diametric opposition to what is needed for adequate defense of a claim of false confession. Dicks attorney, Michael Fasanaro, for example, had a flat fee arrangement with Dicks father for $22,500, with $500 designated for investigationa woefully inadequate budget for adequate defense of a capital case or a claim of false confession, and an arrangement in conflict with the position of the American Bar Association that flat fees are inappropriate in capital cases (American Bar Association, 2003). VI. Inadequate context for evaluation of evidence: Inadequate and/or misleading relevant knowledge and beliefs Clearly, to avoid false confessions in the first place, or to successfully detect them when they do occur, it is crucial to understand their existence and the manner in which they are caused. Without this knowledge, police are likely to unknowingly engage in the very behaviors that cause false confessions. And, firmly convinced that false confessions (particularly to serious crimes) simply do not occur among sane suspects of normal intelligence, they and others to follow are unlikely to even consider the possibility that a confession is false, or to engage in follow-up investigations that would verify or discredit the confession. Likewise, if they lack relevant knowledge, defense attorneys are also likely to

43

disbelieve claims of innocence, and to be unable or unwilling to successfully mount a false confession defense. Indeed, these are the patterns we have outlined above. In the Norfolk Four case, the defense attorneys for Williams, Dick Wilson and Tice lacked experience litigating false confession cases. Lack of relevant knowledge concerning false confessions is pervasive among those who generate and judge them. Police receive poor and misleading training about the risks of psychological interrogation and police-induced false confessions (Leo & Ofshe, 1998; Davis & ODonohue, 2004). Interrogation manuals have widely neglected the subject of false confessions, despite considerable published research documenting their existence and effects (Kassin & Gudjonsson, 2004; Drizin & Leo, 2004; Gudjonsson, 2003; Davis & ODonohue, 2004; Leo & Ofshe, 1998; Ofshe & Leo, 1997a, 1997b). The widely cited Inbau and Reid manual, for example, did not discuss the problem of policeinduced false confessions until its fourth edition in 2001, when a chapter on the subject was added. Still, like every other American interrogation manual and training program the authors continue to insist that the methods it advocates are not apt to lead an innocent person to confess (Inbau et al., 2001, p. xvi). As a result, many American police not only fail to understand the causes and effects of police-induced false confessions, but they also possess incorrect and misleading beliefs based on their training. Some, such as erroneous beliefs about what behaviors reflect deception, lead them to inaccurately classify anxiety as deceptionand therefore to misinterpret innocent behavior as reflecting guilt (Kassin & Gudjonsson, 2004; Vrij, 2008). Others, such as denial that prominent interrogation methods can induce false confessions, lead them to use them without concern, and to believe in the accuracy of the confessions they produce. For example, some forms of minimization, (called theme development) which is one of the cornerstones of the Reid technique (Inbau et al., 2001), have

44

FALSE CONFESSION

clearly been shown to convey expectations of leniency, and to be implicated as causes of false confession in both real-world case studies and laboratory experiments (see Leo, 2008; Kassin & Gudjonsson, 2004, for reviews). Yet, the Inbau manual continues to deny that theme development either conveys promises of leniency or would lead an innocent person to confess. Such erroneous beliefs may lead police to fail to question the validity of the confessions they obtain, or to fail to follow up with further investigation or careful comparison of the details of the suspects postadmission narrative to the evidence. Similar misleading or absent knowledge also appears to be pervasive among criminal defense attorneys. Although there is no published data on what training defense attorneys receive about actual police interrogation practices, both authors have worked with many criminal defense attorneys who have never read an interrogation manual, attended an interrogation course, or been exposed to important reviews of the scientific literature on interrogation and confession. Many have also not received training or instruction on how to defend a false confession case, as was appeared to be the case for the attorneys for Wilson and Tice. Such training is especially important for defense attorneys, particularly those who represent defendants accused of murder, which was the crime in question for approximately 80% of proven false confessions (Drizin & Leo, 2004). Such training should incorporate issues of tunnel vision and confirmation biases, since defense attorneys often inappropriately rely solely on selective evidence generated by detectives guided by such biasesor worse, on restricted evidence truncated when the confession stopped further investigation. It should also cover vulnerabilities to tunnel vision in the defense attorney. As occurred in the Norfolk Four case, defense attorneys may fail to adequately consider and pursue exoneration, focusing instead on the sole goal of avoiding execution or otherwise minimizing sentencing.

45

Similar problems of knowledge occur in trial judges, juries, and appellate judgesperhaps most fully among jurors, who have even less exposure to the phenomenon of false confession than those in the legal system (Chojnacki, Cicchini, & White, 2008). Thus, expert testimony may become an important mechanism of education for judges and factfinders (Costanzo & Leo, 2007). The impact of the dearth of knowledge about false confessions can be exacerbated by other misleading or missing knowledge. That is, other failures of knowledge can lead the confession to appear more valid. Prominent among these are inaccurate beliefs regarding diagnostic cues to deception and guilt. Although humans have been shown in a large scientific literature on detection of deception to perform at no better than chance levels (see Vrij, 2008 for review), they tend to believe they can accurately diagnose deception. And, although several studies have shown that observers cannot differentiate true from false confessions (see Kassin & Gudjonsson, 2004 for review), jurors and others in the justice system are faced with just such a task. A number of false beliefs affect these judgments, in addition to the simple lack of belief in false confession. These can include the idea that even if some do falsely confess, it is only the retarded or insanean idea clearly disproven by Tice (whose IQ was measured at 148-164 and who was not mentally ill)but one that certainly contributed to doubt that his and the others confessions could be false. Indeed, the published data indicate that most false confessions are given by mentally normal, not insane or cognitively impaired, adults (Drizin & Leo, 2004; Leo & Ofshe, 1998; Kassin & Gudjonsson, 2004; Davis & ODonohue, 2004). Other damaging misleading beliefs concern the nature of what could and could not be included in a confession if it were false. Earlier we discussed the damning impact of confession narratives that include motives, detailed crime knowledge, and acknowledgements of voluntariness that

46

FALSE CONFESSION

would, on its face, seem unlikely if the person did not commit the crime. In addition, jurors may make much of displays of emotion without understanding of their actual causes. Tice, for example, sounded very cold and emotionless as he recounted a horrific gang rape and murder, leading observers to conclude that he was cold blooded and possibly psychopathic. Dick cried on the witness stand as he described the incident. He says he cried because he believed he recollected Michelles dead body as it looked in the photos, even as observers concluded (wrongly) that it meant he felt remorse for his supposed role in her murder. The case was replete with mistaken interpretations of the actions of the defendants, based on mistaken lay theories about what actions must mean in particular contexts (e.g., what constitutes an unlikely or inappropriate emotion in what context), typically serving to confirm expectations of guilt. VII. Inadequate context for evaluation of evidence: Progressively constricting relevant information Those who must judge the validity of a confession face constricting potential for exposure to evidence that would support a claim of false confession over time. That is, the sum of available evidence becomes progressively more selective as it becomes filtered through prior actorsas, for example, when jurors must rely on what is presented to them, as opposed to all evidence collected by all parties. To the extent that initial collection of evidence is constricted by selective search and attention, the availability of disconfirming evidence can be impaired from the very beginning of the investigation, sometimes in ways that cannot be corrected. When attention prematurely focuses exclusively on one suspect, for example, the existence of other suspects may never be investigated or recognized, leaving relevant evidence to be ignored, lost, or contaminated. Particularly relevant for claims of false confession is all evidence regarding the initial interactions of police with the suspect, all evidence relevant to probable cause for targeting the suspect (including leads to other suspects), and all evidence regarding the interrogation itself. When

47

interrogations are unrecorded, fact finders must rely on the original notes of the interrogators and records of the suspects memories and reports to others. Unfortunately, such notes are often selective, and may be lost or destroyed before they can be challenged at trial or in pre-trial proceedings. For example, after taking Williams taped confession,
Halverson and Evans would have their handwritten notes of the interrogation typed up, cleaned up to their advantage in the process, then perhaps cleaned up some more; later those handwritten notes might be thrown away so that defense attorneys could not use them as a basis for challenging what had transpired in the interrogation room. That was standard operating procedure for Norfolk detectives. . . . (Wells & Leo, 2008, p. 33)

The juries in the trials of Wilson and Tice may well have reached different verdicts had videotapes of the defendants interrogations existed and had these juries been able to view them. Moreover, even where a full recording of an interrogation is available, the full context includes not only the interrogations techniques, but also timing of the interrogation, and the condition of the suspect (e.g., sleep deprivation, intoxication, nutrition, recent meals, emotional stress, etc.), factors which can leave a suspect more vulnerable to police pressure. To the extent that prosecutors and defense attorneys rely on police investigative reports as the sole source of evidence, they necessarily receive less information than that originally available to police (see Findley and Scott, 2006 for discussion of selective provision of evidence by police). And, even if they follow up with further investigation, their efforts can be selectively driven by what did appear in police reports, with original potential evidence or leads now lost or unlikely to be recognized (Fisher, 1988). Jurors and later appellate courts must rely on what is presented in trial or briefs, and therefore suffer greater constriction in context for evaluating the evidence they do

48

FALSE CONFESSION

see. Even among the evidence available to them, prosecutors are unlikely to focus attention on exculpatory evidence in trial presentations or appellate briefs, while defense attorneys may have failed to receive such evidence, and therefore be unable to present it. Defense attorneys may also be unaware of how to evaluate the potential for false confession or may be focused on avoiding the death penalty rather than on exonerationand therefore may fail to present full contextual evidence relevant to guilt. Together, strong beliefs that false confessions are rare, if not unknown, combined with lack of understanding of the causes of false confessions, and constricted evidence for evaluation of claims of false confession present formidable obstacles for exoneration before or at trial, or for post-conviction relief. Research on attribution of causes of behavior has shown that observers are biased toward making internal attributions for why a person behaves in a particular way. That is, observers tend to assume that behavior is voluntary and that it reflects the wishes and dispositions of the actor, even when available contextual information includes plausible external causes of behaviora tendency dubbed the fundamental attribution error ( Ross, 1977) or the correspondence bias (Jones, 1979). This bias toward internal attributions for behavior is so strong that in one study, for example, observers made such internal attributions (assuming that the persons actions reflected his actual feelings) even when the target was given no choice of which way to behave (Jones & Harris, 1967). By definition, the correspondence bias reflects our tendencies to fail to notice or to discount external forces affecting others behavioral choices. When brought to bear on judgments of the validity of a confession, this natural bias is enhanced by the assumption that people will not confess to heinous crimes they did not commit, and by the lack of ability to recognize the contextual forces that would promote a false confession (coercive interrogation methods, the vulnerabilities and condition of the suspect, etc.), and by the

49

lack of available information regarding them (e.g., lack of recordings of the interrogation or records concerning the condition of the suspect). Nevertheless, understanding of the full context in which a confession takes place can only benefit judges and juries attempting to evaluate claims of coercion and invalidity. The importance of even small details of context has been demonstrated in a line of research by Daniel Lassiter and his colleagues (for a review see Lassiter, Ware, Lindberg, & Ratcliff, in press) in which they have shown that even the angle from which an interrogation is recorded can affect judgments of coercion and validity. That is, confessions tend to be judged as more valid and voluntary when the camera focuses only on the suspect than when it focuses equally on both suspect and detective, and least voluntary when it focuses only on the detective. Thus, being unable to see the full context, including both participants, tends to lead observers to focus on the causal role of the one they can see, without full appreciation of the role of the other. If such a small aspect of the contex in which a confession takes place, when missing, can significantly increase judgments of voluntariness and validity, we can only imagine the magnitude of the effect of all missing contextual details. Conclusion To date the social science literature on false confessions has primarily addressed three issues: the psychological methods and processes of interrogation that lead to false confession, individual susceptibilities to interrogative influence and false confession, and the role that false confessions have played in documented cases of wrongful conviction. The psychological processes through which false confessions lead to wrongful conviction have been largely unaddressed. This article has taken up this challenge, addressing this gap in the literature by examining seven psychological processes linking false

50

FALSE CONFESSION

confession to wrongful conviction and failures of postconviction relief. For those interested in preventing miscarriages of justice, it is imperative to study more closely the link between false confession and wrongful conviction. On the one hand, there is no piece of erroneous evidence that if put before a jury is more likely to lead to a wrongful conviction than a false confession (Kassin & Neumann, 1997; Leo & Ofshe, 1998). Criminal justice officials and jurors treat confessions as the crme de la crme of prosecutorial proof (Davies, 2006), and as a result false confessors whose cases go to trial are convicted the vast mjority of the time (Leo & Ofshe, 1998; Drizin & Leo, 2004). On the other hand, not all false confessions lead to the wrongful conviction of the innocent: Sometimes police do not bring confession cases to prosecutors, sometimes prosecutors drop charges, sometimes judges suppress confessions, and sometimes juries acquit false confessors. The key to preventing confession-based miscarriages of justice is therefore to better understand why some false confessions lead to wrongful convictions and others do not. This article is a first step in that direction. Notes
1. Williams had not failed the polygraph at all. His physiological responses to the test questions, in fact, indicated that he had passed the polygraph. The idea that memories can be repressed has been thoroughly discredited in the scientific research literature. For recent reviews, see Davis & Loftus (in press); Piper, Lilevik and Kritzer (2008); and Loftus and Davis, 2006.

2.

References

American Bar Association. (2003). American Bar Association guidelines for the appointment and performance of defense counsel in death penalty cases. Washington, DC: Author. Bedau, H. A., & Radelet. M. L. (1987). Miscarriages of justice in potentially capital cases. Stanford Law Review, 40, 21-179.

51

Cassell, P., & Hayman, B. (1996). Police interrogation in the 1990s: An empirical study of the effects of Miranda. UCLA Law Review, 43, 839-931. Castelle, G. & Loftus, E. (2001). Misinformation and wrongful convictions. In S.D. Westervelt & J.A. Humphrey (Eds.), Wrongly convicted: Perspectives on failed justice (pp. 17-35). Newark: Rutgers University Press. Chojacki, D.E., Cicchini, M.D., & White, L.T. (2008). An empirical basis for the admission of expert testimony on false confessions. Arizona State Law Journal, 40, 1-45. Colorado v. Connelly, 479 U.S. 157 (1986). Connery, D. (2008). Foreword. In T. Wells and R. A. Leo (Eds.), The wrong guys: Murder, false confessions and the Norfolk 4, (pp. ix-xi). New York: The New Press. Costanzo, M., & Leo, R.A. (2007). Research and expert testimony on interrogations and confessions. In M. Costanzo, D. Krauss, & K. Pezdek (Eds.), Expert psychological testimony for the courts (pp. 69-98). Mahwah, NJ: Lawrence Erlbaum Associates Publishers. Davies, S. (2006). The reality of false convessionslessons of the Central Park Jogger case. New York University Review of Law and Social Change, 30, 209-253. Davis, D., & Follette, W. (2002). Rethinking probative value of evidence: Base rates, intuitive profiling and the postdiction of behavior. Law and Human Behavior, 26, 133-158. Davis, D., & ODonohue, W. (2004). The road to perdition: Extreme influence tactics in the interrogation room. In W. ODonahue (Ed.), Handbook of forensic psychology (pp. 897-996). San Diego: Academic Press. Davis, D., & Leo, R. A. (2006). Strategies for preventing false confessions and their consequences. In M. Kebbell and G. Davies (Eds.), Practical psychology for forensic investigations and prosecutions (pp. 121-149). New York: John Wiley & Sons. Davis, D., & Friedman, R. D. (2007). Memory for conversation: The orphan child of witness memory researchers. In M. P. Toglia, J. D. Read, D. F. Ross, & R. C. L. Lindsay (Eds.), The handbook of eyewitness psychology, Vol. I: Memory for events (pp. 3-52). Mahwah, NJ: Lawrence Erlbaum Associates.

52

FALSE CONFESSION

Davis, D., Weaver, T., & Leo, R. A. (2007, August). Effects of failed polygraph results on true and false confessions. Paper presented at the meeting of the American Psychological Association, San Francisco, CA. Davis, D., Leo, R. A., & Follette, W. C. (2008, March). Recommending false confession for the innocent. Paper presented at the meeting of the American Psychology-Law Society. Ft. Lauderdale, FL. Davis, D., & Loftus, E. F. (2009). Expectancies, emotion and memory reports for visual events. In J. R. Brockmole (Ed.), The visual world in memory (pp. 178-212). New York: Psychology Press. Deffenbacher, K. A., Bornstein, B. H., Penrod, S. D., & McGorty, E. K. (2004). A meta analytic review of the effects of high stress on eyewitness memory. Law and Human Behavior, 28, 687-706. Drizin, S., & Leo, R. A. (2004). The problem of false confessions in the post-DNA world. North Carolina Law Review, 82, 891-1007. Easterbrook, J. A. (1959). The effect of emotion on cue utilization and the organization of behavior. Psychological Review, 66, 183-201. Elliot, A. J., & Dweck, C. S. (2005). Handbook of competence and motivation. New York: Guilford Publications. Festinger, L. (1957). A theory of cognitive dissonance. Oxford: Row, Peterson. Findley, K., & Scott, M. (2006). The multiple dimensions of tunnel vision in criminal cases. Wisconsin Law Review, 2006, 291-398. Finkel, N. J., & Parrott, W. G. (2006). Emotions and culpability: How the law is at odds with psychology, jurors, and itself. Washington, DC: American Psychological Association. Fisher, S. (1988). In search of the virtuous prosecutor: A conceptual framework. American Journal of Criminal Law, 15, 197-261. Follette, W. C., Davis, D., & Leo, R. A. (2007). Mental health status and vulnerability to interrogative influence. Criminal Justice, 22, 42-49. Gilovich, T. (1991). How we know what isnt so: The fallibility of human reason in everyday life. New York: The Free Press. Givelber, D. (2000). Punishing protestations of innocence: Denying responsibility and its consequences. American Criminal Law Review, 37, 1363-1408.

53

Gross, S., Jacoby, K., Matheson, D., Montgomery, N., & Patel, S. (2005). Exonerations in the United States, 1989 through 2003. Journal of Criminal Law and Criminology, 95, 523-553. Gudjonsson, G. H. (2003). The psychology of interrogations and confessions: A handbook. New York: Wiley. Hammond, K. R. (2000). Judgments under stress. New York: Oxford University Press. Hassin, R. R., Uleman, J. S., & Bargh, J. A. (2005). The new unconscious. New York: Oxford University Press. Inbau, F. E., Reid, J. E., Buckley, J. P., & Jayne, B. C. (2001). Criminal interrogations and confessions (4th ed.). Gaithersburg, MD: Aspen. Jones, A., & Fitness, J. (2008). Moral hypervigilance. Emotion, 8(5), 613-627. Jones, E. E. (1979). The rocky road from acts to dispositions. American Psychologist, 34, 107-117. Jones, E. E., & Harris, V. A. (1967). The attribution of attitudes. Journal of Experimental Social Psychology, 3, 1-24. Kassin, S., & Fong, C.T. (1999). Im innocent! Effects of training on judgments of truth and deception in the interrogation room. Law and Human Behavior, 23, 499-516. Kassin, S., Goldstein, C.J., & Savitsky, K. (2003). Behavioral confirmation in the interrogation room: On the dangers of presuming guilt. Law and Human Behavior, 27, 187-203. Kassin, S., & Gudjonsson, G. (2004). The psychology of confessions: A review of the literature and issues. Psychological Science in the Public Interest, 5, 35-67. Kassin, S., & Neumann, K. (1997). On the power of confession evidence: an experimental test of the fundamental difference hypothesis. Law and Human Behavior, 21, 460-484. Kassin, S., & Sukel, H. (1997). Coerced confessions and the jury. Law and Human Behavior. 21, 27-46. Kassin, S. M., & McNall, K. (1991). Police interrogations and confessions: Communicating promises and threats by pragmatic implication. Law and Human Behavior, 15, 233-251.

54

FALSE CONFESSION

Lassiter, G. D., Ware, L. J., Lindberg, M. J., & Ratcliff, J. J. (in press). Videotaping custodial interrogations: Toward a scientifically based policy. In G. D. Lassiter & C. A. Meissner (Eds.), Interrogations and confessions: Current research, practice and policy. Washington DC: American Psychological Association. Law and Human Behavior. (2006). Special issue: Emotion in legal judgment and decision making, 30(2). Leo, R. A. (1996). Inside the interrogation room. Journal of Criminal Law and Criminology, 86, 266-303. Leo, R. A. (2008). Police interrogation and American justice. Cambridge, MA: Harvard University Press. Leo, R. A., & Ofshe, R. (1998). The consequences of false confessions: Deprivations of liberty and miscarriages of justice in the age of psychological interrogation. Journal of Criminal Law and Criminology, 88, 429-296. Leo, R. A., & Ofshe, R. (2001). The truth about false confessions and advocacy scholarship. The Criminal Law Bulletin, 37, 293-370. Leo, R. A., Drizin, S., Neufeld, P., Hall, B., & Vatner, A. (2006). Bringing reliability back in: False confessions and legal safeguards in the twenty-first century. Wisconsin Law Review, 2006, 479-539. Levine, L. J., & Pizzarro, D. A. (2004). Emotion and memory research: A grumpy overview. Social Cognition, 22, 530-554. Martin, D. (2002). Lessons about justice from the laboratory of wrongful convictions: Tunnel vision, the construction of guilt, and informer evidence. UMKC Law Review, 70, 847-864. Medwed, D. (2004). The zeal deal: Prosecutorial resistance to postconviction claims of innocence. Boston University Law Review, 84, 125-183. Meissner, C., & Kassin, S. (2002). Hes guilty! Investigator bias in judgments of truth and deception. Law and Human Behavior, 26, 469-480. Meissner, C. A., & Kassin, S. M. (2004). Youre guilty, so just confess?! Cognitive and behavioral confirmation biases in the interrogation room. In G. D. Lassiter (Ed.), Interrogations, confessions and entrapment (pp. 85-106). New York: Kluwer Academic.

55

Miller, G. R., and Boster, F. J. (1977). Three images of the trial: Their implications for psychological research. In B. Sales (Ed.). Psychology in the Legal Process (pp. 19-38). New York: Pocket Books. Nardulli, P., Eisenstein, J., & Fleming, R. (1988). The tenor of justice: Criminal courts and the guilty plea process. Urbana: University of Illinois Press. Ofshe, R., & Leo, R.A. (1997a). The social psychology of police interrogation: The theory and classification of true and false confessions. Studies in Law, Politics and Society, 16, 189-251. Ofshe, R., & Leo, R. A. (1997b). The decision to confess falsely: Rational choice and irrational action. Denver University Law Review, 74, 979-1122. Psychological Inquiry. (1999). Special issue on motivated cognition, 10(1). Redlich, A. (In Press). False confessions, false guilty pleas: Similarities and differences. In D. Lassiter & C. Meissner (Eds.), Interrogations and confessions: Current research, practice and policy. Washington, DC: American Psychological Association. Ross, L. (1977). The intuitive psychologist and his shortcomings: Distortions in the attribution process. In L. Berkowitz (Ed.), Advances in experimental social psychology Vol. 10, (pp. 174221). New York: Academic Press. Safer, M. A., Christianson, S.-., & Autry, M. W. (1998). Tunnel memory for traumatic events. Applied Cognitive Psychology, 12(2), 99-117. Scheck, B., Neufeld, P., & Dwyer, J. (2000). Actual innocence: Five days to execution and other dispatches from the wrongly convicted. New York: Random House. State of California v. Cahill, 5 Cal. 4th (1993). Tavris, C., & Aronson, E. (2007). Mistakes were made (but not by me): Why we justify foolish beliefs, bad decisions and hurtful acts. New York: Harcourt. Thomas, G. (2008). The Supreme Court on trial: How the American justice system sacrifices innocent defendants. Ann Arbor: University of Michigan Press.

56

FALSE CONFESSION

Vrij, A. (2008). Detecting lies and deceit: Pitfalls and opportunities. (2nd ed.). London: Wiley. Warden, R. (2003). The role of false confessions in Illinois wrongful murder convictions since 1970. (Center on Wrongful Convictions Research Report). Retrieved December 1, 2009 from www.law .northwestern.edu/depts/clinic/wrongful/FalseConfessions.htm Wells, T., & Leo, R. A. (2008). The Wrong guys: Murder, false confessions, and the Norfolk Four. New York: New Press.

Copyright of Journal of Psychiatry & Law is the property of Federal Legal Publications Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.