Wrongful Convictions Reading List

Vitrol & Koreva, 2018

Capital voir dire CVD (including death qualification) led a significant belief in suspect guilt before a trial yet led to lower conviction rates as compared to standard voir dire processes SVD Study 1: (2x3) CVD x SVD; weak x moderate x strong evidence... Main effects found, no interactions... CVD condition significant difference in suspect guilt pretrial, likely to favor DP. Significant differences in conviction rates regarding evidence type where moderate strength of evidence having higher higher rates than weak evidence, and strong evidence having higher conviction rates vs moderate evidence. These pretrial guilt ratings didn't "survive" the trial evidence phase, CVD less likely to convict. Study 2: CVD vs SVD for required evidence (the type and amount of inculpatory evidence needed to convict). CVD required significantly more evidence, and specifically physical evidence, information on motives and background of the defendant. CVD expected more evidence presented needed to reach verdict Still need more research on underlying mechanisms involved in determining guilt

Kassin et al., 2005

DECEPTION DETECTION College students and police investigators watched or listened to 10 prison inmates confessing to crimes. Half the confessions were true accounts; half were false—concocted for the study. Consistent with much recent research, students were generally more accurate than police, and accuracy rates were higher among those presented with audiotaped than videotaped confessions. In addition, investigators were significantly more confident in their judgments and also prone to judge confessors guilty. To determine if police accuracy would increase if this guilty response bias were neutralized, participants in a second experiment were specifically informed that half the confessions were true and half were false. This manipulation eliminated the investigator response bias, but it did not increase accuracy or lower confidence. These findings are discussed for what they imply about the post-interrogation risks to innocent suspects who confess.

Leach, 2012

DECEPTION DETECTION 100% of wrongful conviction cases the truth was overlooked Labeling a lie teller sets the stage for conviction Legal system participants may be highly motivated to lie Accomplices, jailhouse informants Police (intentional acts) HISTORY Hugo Munsterberg (1915) made note of importance of deception detection Deception Detection Interpersonal communication Person perception Decision making Discrimination between liars and truth tellers Bias THEORIES Emotion Physiological patterns and cognition Microexpressions, creation of lie Self-presentation perspective Focus on masking the lie and presenting cues that indicate truth telling Cognition Increasing cognitive load Lying more cognitively demanding Cognitive Dissonance Theory Discrepancy between cognition and Behavior Lying is wrong, telling a lie METHODS Lie production paradigms Individuals either tell the truth or lie Ground truth is known Either about opinions, life events, or witnesses events Criticized for lack of external validity/realism Two issues Low vs high stakes No reward/consequence; too high unethical Exp manipulated vs naturalistic deception Told to lie, "an actor is not a liar"; must lie on their own accord, hard to manipulate ethically There is no gold standard; each method has its pitfalls Lie Detection Paradigms Can occur during the lie/truth (online) or after a delay Approaches can vary in training and equip required Can focus on specific cues indicative of deception or global judgements Polygraph Monitoring physiological arousal can reveal deception Some Polygraphs rely on guilty knowledge paradigm Info only the perp should know - should produce orienting response Better theoretical basis BUT more false negatives BUT can't distinguish if they know because they did it or because they were exposed to info somewhere Often used for purposes other than intended Lie, tell suspect they failed regardless Use as a tell-all; in pre-polygraph interview, hope that people reveal information out of fear of future failure of polygraph Overall, can accurately discriminate between liars and truth tellers BUT high # false positives Methods exist to counter, falsify patterns Neuroimaging techniques fMRI Event-related potentials Still being developed, research is limited Relies on guilty knowledge paradigm Statement Analysis Analysis of verbal content Criteria-based content analysis (CBCA) Indicators of truth, no deception Based on transcripts only (verbal, not B) First step in SVA (statement validity analysis) Reality Monitoring (RM) Similar to CBCA True stories will include reality checks (did this happen like this; providing sensory details; spontaneous corrections) Also generally indicators of truth Better than chance detection, but high training and time consuming Intuitive Deception Detection Most common, direct approach Asked to view a video, make a decision whether liar or truth teller RELEVANT RESEARCH Differences exist between liars and truth tellers Verbal and B cues However, no single cue Must be readily apparent to observers Deception detection accuracy Average layperson detects at 54%, near chance levels LEOs perform the same though Very few groups are better than chance Majority of characteristics studies have no effects on deception detection Lie detection performance not reliable under various conditions Might just be guessing Accuracy and confidence Low accuracy, high confidence BUT no relationship LEOs are more confidence in their abilities than laypersons Training Reid = Behavior Analysis Interview (BAI) Cues trained to look for are NOT empirically indicative of deception Basis of BAI (truth tellers will score higher) is incorrect Liars concerned with impression management (helpful, less nervous) High Reid Cues associated with lower levels of lie detection accuracy Reid bolsters confidence and reduces accuracy Application Research has little impact on policy Common occurrence Challenging for empirically based training programs to compete with Reid Claims 85% success rate Research can't top that Two main ways to improve accuracy Making deception more difficult for liars Cognitive load Improving lie detection techniques SUE Best way: introduce caution; tell people how bad they are, despite their confidence

Masip et al., 2011

DECEPTION DETECTION Common sense Hard to justify, but generally assumed reliable Commonsense psychology Antecedents and consequences of behavior Heuristic processing and defective strategies Differs from experimental psychology Goal of paper: to show that the BAI is a common-sense technique marketed as a superior tool, despite empirical evidence stating otherwise BAI Police officers are at 55% accuracy for deception detection Interview protocol aimed at generating observable differences between liars and truth tellers Background questions Investigative questions Behavior-provoking questions Less comfortable Less helpful Less nervous about being a suspect Empirical research on Reid Criminal Interrogations and Confessions Book Behavioral indicators of deception Only 3 of 9 empirically associated with veracity; one of them associated in opposite direction (pitch) When they are significantly effective, they are so in the opposite direction Found that deception detection was LOWER after being trained in cues Overall, B cues of Reid DO NOT correlate with empirical research BAI Only 3 empirical studies to date One with trained interviewers and BAI experts but NO SEPARATE GROUND TRUTH 2 of the authors were authors of the BAI Sample size: 4 BAI experts Second with students but lots of methods problems 7 correct in control condition, 8 correct in BAI condition Third (vrij) found either no effects or effects opposite results of BAI Nine step interrogation method Been shown to be psychologically coercive and yield false confessions Present Study Extend Vrij (2006) findings that BAI indicators are wrong To see if BAI is just intuitive method of lie detection Study 1 Test if BAI recommendations are just common sense Accuracy of guilty suspect detection Findings 7/10 correct IDs in control BUT BAI condition almost 100% accuracy Limitations 2 interviews were used Might have been prototypical, not typical Some presented as written text Study 2 Assess laypersons beliefs about verbal and non-verbal correlates of guilt and innocence during an interview Findings People who did not know BAI rated guilt indicators as highly deceptive compared to innocence indicators Confirms BAI is just common sense Implications BAI training should not be included in Reid Seminars - people already know it intuitively If BAI indicators are intuitive, then actual suspects may know them already and actively try to avoid displaying guilty indicators Putting innocent suspects at risk

Zimmerman et al., 2012

EYEWITNESS TESTIMONY APPROACH Chapter on Suggestive Eyewitness ID procedures MAIN FINDINGS Main sources of system error Lineup instructions Suspect is here, you must pick him TP vs TA changes how effective unbiased instructions are Lineup composition Who's in the lineup? Who are the fillers? Are the photos of similar quality? Size? Color vs black and white? Type of photo (selfie vs mugshot) Use a functional/effective group size Admin knowledge of suspect ID Behavioral cues Single blind only Lineup presentation method Single suspect ("showups") Simultaneous lineups Multiple presentations of suspect Leads to transference Outline of most eyewitness experiments Witness staged event, participate in some form of identification procedure APLS Recommendations (1998) Unbiased instructions Perpetrator may or may not be in lineup Admin does not know suspect identity Double blind procedure Admin should not know suspect identity Plausible fillers Lineup composition shouldn't make suspect stand out Similar to witness description, photos all of similar quality Get decision confidence rating prior to admin feedback/reaction Didn't officially favor sequential over simultaneous NIJ Recommendations (1999) Unbiased instructions Includes warning that suspect may have changed appearance Unsupported by psych studies! Supposedly increases false IDs, lowers criteria to choose Did not recommend double blind Did not recommend sequential over simultaneous IMPORTANT TAKEAWAYS To avoid a biased or suggestive eyewitness ID, use: Unbiased instructions Plausible fillers Double blind procedure Sequential lineups better at reducing false identification Avoid repeat presentations

Fulero, 2010

Expert Testimony Admission Criteria Frye Rule Whether the expert's opinion was developed using methods generally accepted in the relevant professional community? Daubert Rule Can the new ideas be tested? Have they been subject to peer review? Are the methods generally accepted? Are there rates of errors and generally accepted standards for the technique? Burden on Defense Defense must link expert testimony to facts of the case Cannot address specifics of the crime Unless clinical, then can speak to mental state of specific witness or defendant Criticisms Not all courts will accept expert testimony Doesn't provide information outside general knowledge Not applicable, no evidence exists in trial that warrants expert Without tapes, there often isn't proof of coercion/false confession Five main arguments Against Not scientific or not reliable and valid Needs to be conducted via accepted and known methods of scientific research Needs to be in an area of generally accepted scientific research Surveys of experts Coverage in textbooks Peer-reviewed # of citations Coverage in graduate programs APA Needs a known or potential error rate Not always obtainable or needed, ridiculous Needs to be testable Invades domain of the jury Informational, not influential Within the knowledge of the average layperson Underestimate power of information Can all be done in Cross-Examination Leaves defense with no proof, therefore no case Probative value vs prejudicial effect Max vs min is the goal Conclusions Successful testimony doesn't enter the record: need a database of all expert testimony "...it is for the expert to lead the jurors to water, but it is for the attorneys to make them drink." Pg 221

Chapter 2 Garrett, 2011

FALSE CONFESSIONS APPROACH Overview chapter of contaminated false confessions, specifically coerced-compliant confessions METHOD Police Record parts (if any) of interrogation Confession evidence stops investigation Court Confessions appear accurate Experts appear less accurate MAIN FINDINGS Reliable Confessions: Must have been advised of Miranda Must have been Voluntary Factors that deem involuntary: Youth Low Intelligence No Miranda Length/Type of questioning Physical threats Demeanor during questioning IMPORTANT TAKEAWAYS Ideas for reform Mandatory taping of full interrogations Reforming criminal procedure

Collins and Jarvis, 2009

FORENSIC SCIENCE Review Garrett/Co-founder of innocence project's claim that faulty forensics are main cause of wrongful convictions Examined first 200 cases (same cases evaluated by Garrett) Background Edwin Borchard (law prof) first to examine wrongful convictions and propose reforms Courts should not introduce confession until it's given before magistrate and witnesses Independent investigative bodies should review wrongful convictions 1989 kicked innocence advocates off Ted Bundy executed, then Gary Dodson first to be exonerated by DNA 1992 Innocence project created by defense attorneys Scheck and Neufeld 2007 Garrett reports 113/200 cases involved forensics, and 57% had misleading testimony Media frenzy ensues Neufeld pointed finger at crime labs Kept up provocative statements during major forensic conferences/conventions until Senate convened hearing to address alleged failure of Justice Department 2008 Hearing "Prosecution" Unreliable/unvalidated science and/or misconduct in forensics second highest contributor to wrongful convictions "Defense" Correlation ≠ Causation Garrett study about leading types of evidence in wrongful convictions, not that evidence was 2nd contributor Investigators (mainly Neufeld) have been LAWYERS Trained in cause and effect No scientific objective approach, no consideration of confounds or alternative hypotheses Evidence 97 cases involved forensic science (328 instances of error; used this number for most percentages) Only 27% of those were systematic errors due to forensic science 36/200 (18%) had forensic science favorable to defendant Yet blamed on unreliable science Because forensic evidence did not prevent wrongful conviction, it was deemed faulty by Innocence Project But evidence that cannot exclude the defendant is fair game at trial Malpractice Termed faulty forensics by public May actually be valid/reliable methods Public doesn't understand what is real forensic science Definition "the failure of a professional person to act in accordance with the prevailing professional standards, or failure to forsee consequences that a professional person, having the necessary skills and education, should forsee" (p. 24) "Closing Arguments" Bad Lawyering/Government Misconduct Innocence Project stated that 3/200 cases were due to bad lawyering This study suggested that almost all cases now exonerated could have been prevented with better lawyers on both sides Example: Krone "Snaggle Tooth Killer" Science was ignored or redone to get results Blamed attorneys for not finding this Forensic science malpractice does occur But 73% of crime labs now accredited with strict oversight Innocence movements assume: Faulty forensics leading cause of wrongful convictions Crime lab accreditation fails to minimize/prevent malpractice Oversight commissions by government will address issues of integrity BUT ALL INVALID Conclusion Forensic science factors are not the causes of wrongful convictions, but symptoms of a bigger problem of injustice

Pezdek, 2012

Fallible Eyewitness Memory and Identification Overview chapter of (a) scientific methods for studying eyewitness identification evidence, (b) relevant research on eyewitness memory, and (c) applications of eyewitness research Main findings (A) Scientific methods for studying EW identifications Case Studies In-depth study of one individual or incident Yuille and Cutshall (1986): Analyzed verbatim reports given by witnesses from British Columbia (21 immediately after; 13 again after delay) who saw a thief break into a gun shop, tie up the owner, and steal several arms/money; As the thief ran out, the store owner freed himself and shot/killed thief Found that witness descriptions were accurate and exhibited little change across retention interval Reported that stress was unrelated to memory BUT level of witness stress and degree of involvement in the incident were confounded, so this conclusion is more or less invalidated Limitations of case study research: (1) Effect of specific eyewitness conditions can rarely be isolated (conditions tend to be confounded with other contextual factors), and (2) Single event can only contain limited set of conditions and a limited number of eyewitnesses Generalizability from one case study to another is highly limited Archival Methods Examining results across many real-world cases Limitation: Cases are only of value if the ground truth is known (and this is rarely the case) Behrman and Davey (2001): Analysis of 271 actual police cases from Sacramento, CA; went through and categorized the conditions under which suspect identifications were more or less likely to occur Cases in which the suspect (who may/may not have been the perpetrator) was selected by the witness BUT ground truth was known for very few cases and some of the findings reported were in direct opposition with what is seen in other parts of research literature E.g., suspect IDs from photo lineups were higher for cross-race than same-race IDs Hard to verify accurate suspect IDs in real-world cases, but it is possible to verify filler IDs Wright and McDavid (1996): Analyzed 1,561 lineup decisions that occurred in London; 19.9% of these decisions were filler IDs Behrman and Davey (2001): 24% of live lineup IDs were filler IDs Valentine, Pickering, and Darling (2003): 21.6% of 119 lineup decisions (London) were filler IDs Archival studies are useful for documenting misidentifications but not for understanding conditions under which witnesses are likely to be accurate Other area of archival research = post-conviction DNA exonerations Has documented that mistaken IDs are likely to occur AND, given the data available, allows for some assessment and association of certain factors with inaccurate IDs Experimental Methods Allow researchers to examine effects of specific independent variables on specific dependent variables (allows for causal conclusions); Participants are randomly assigned to conditions which gets rid of confounds Causal conclusions = more accurate predictions Important to use experimental design in eyewitness research Valentine et al. (2003) was an archival study and reported no effect of cross-race IDs, weapon focus, or time delay (all in contrast of the findings of many other studies) Hard to tease apart the various factors that could have had an effect Limitation: ecological validity Differences in witnessing circumstances - BUT much of the experimental research on this does not support the argument that ecological validity is a serious issue O'Rourke, Penrod, Cutler, and Stuve (1989): Tested 'numerous eyewitness factors' across student and non-student samples; consistent findings across populations Bornstein (1999): No main effect of participant sample (students vs. community members) in jury decision making studies Pezdek, Avila-Mora, and Sperry (2010): No differences in mock jurors' perceptions or verdicts as function of trial modality Meta-Analyses Allow for statistical analyses regarding different experimental effects (size and direction, and secondary variables associated with strength) Meissner and Brigham (2001): meta-analysis of cross-race ID effects; Witnesses more likely to correctly ID someone from their own race (compared to someone from another race), and more likely to select the wrong suspect with other-race individuals than same-race individuals Shapiro and Penrod (1986): Meta analysis of 128 identification studies published at that time; conclusions made about which of 19 variables were significant in predicting correct hit rates and false alarms Usually express effect size in terms of d (0 = no effect, anything > 0 indicates better recognition in one condition than another) Limitation: value depends on quality of studies included (B) Relevant Research Estimator variables (event characteristics) Generally true that characteristics of event have more influence on memory than characteristics of eyewitness Exposure time, distance, lighting Far distance, low lighting = weaker memory trace Shorter exposure = greater likelihood of mistaken ID in TA lineup and less likelihood of accurate ID in TP lineup (Memon, Hope, & Bull, 2003) Weapon focus Loftus, Loftus, and Messo (1987): When handgun was present on slides shown to participants, participants made more fixations on the gun; Fixations on gun were longer than fixations on control object (bank check) Subsequent IDs showed that witnesses who had gun present were less likely to ID perpetrator than those who saw check Steblay (1992): Weapon focus effect was greater in TA lineup and when compounded with other conditions Disguise Disguises to upper facial features (eyes, forehead, hairline) more likely to impair ID witness memory than lower feature disguises (mouth, chin, nose) (Davies, Shepherd, & Ellis, 1979) Time delay Shapiro and Penrod (1986): Longer delays led to fewer correct IDs and more mistaken IDs Jost's Law (Britt & Bunch, 1934): Weaker memories fade faster than strong ones - passage of time will have more effect on memory of witnesses whose initial viewing experience was poor Estimator variables (witness characteristics) Confidence Initial reports painted somewhat dismal picture of confidence-accuracy relationship (Sporer, Penrod, Read, & Cutler, 1995); r = .29 but only explained 8% of the variance in accuracy Sporer et al. (1995) showed stronger C-A relationship (r = .41)when limiting analysis to only choosers (this relationship is ore forensically relevant because choosers are the ones who are more likely to to testify in court) #Side note: We know from more updated research that confidence and accuracy are related under "pristine conditions" (Wixted & Wells, 2017) buuuuut this article isn't discussed in the Pezdek chapter (since the chapter was written before the article) Confidence is malleable. Things that inflate EW confidence include: Repeated questioning (Shaw, 1996) or feedback (Wells & Bradfield, 1999) Cross-race IDs One of the strongest witness characteristics associated with ID accuracy (Meissner & Brigham, 2001) Consistent across age (Pezdek, Blandon-Gitlin, & Moore, 2003) Moderating variables: other-race contact (modest), exposure time to target faces Increased viewing time reduces disadvantage for CR faces (Meissner & Brigham, 2001) Stress EW memory is less reliable under high levels of stress (Deffenbacher et al., 2004) Effects recall of crime-related details and person identification ability Effect is greater on reducing correct ID rates than it is on increasing false alarm rates Prosecutors argue that findings of impaired memory under high stress are restricted to lab tests, but empirical research says otherwise (Deffenbacher et al., 2004 - effects of stress were greater for staged crimes than lab-based face recognition tasks; Morgan et al., 2004 - military personnel in prisoner-of-war survival training program showed lower correct ID rate and higher incorrect ID rate for those in high stress condition) Witness intoxication Read, Yuille, and Tollestrup (1992): Alcohol reduced correct IDs, but did not affect false IDs Dysart, Lindsay, MacDonald, and Wicke (2002): False ID rate was higher for intoxicated witnesses BUT this employed a show-up Also remember the Evans and Schreiber Compo papers in which no reliable differences were found, suggesting that alcohol doesn't result in significant memory impairment :) Interactive Effects EW identification accuracy would be expected to be especially unreliable when multiple deleterious factors co-occur (C) Applications of Research on EW Memory Expert testimony Federal Rules of Evidence 702 People v. McDonald (1984): "psychological factors that may have impaired the accuracy of a typical eyewitness ID... with supporting references to experimental studies of such phenomena" Educating defense attorneys and prosecutors Help them discriminate between strong and weak EW evidence in order to avoid proceeding to trial with cases that might lead to a wrongful conviction Important takeaways EW evidence is leading cause of erroneous convictions Research should be disseminated to jurors, prosecution, and defense to reduce wrongful convictions rate

Dror et al., 2006

Forensic Science Experiment on real world fingerprint expert bias from outside contextual information on identification decisions Issue forensic science experts believe they're objective Method Unknowingly tested in their own labs Given a fingerprint they'd matched in a previous case, told that it was not a match but from the Madrid bomber case Asked to confirm/make their own decision Results 4/5 reported that it was not a match Even if had higher sample size, 10% of experts would still be susceptible to misleading contextual information Due to flaw in objectivity in processing/evaluating information Possible sources of contextual effects Emotional context Pressure Contextual information Group think Biases Hopes/expectations Self-fulfilling prophecies Peer pressure Suggestions for future Improve selection and screening of experts Improve training Include methods that can address potential issues along the way Apply relevant cognitive psychology research

Acker & Redlich, 2019

Furman v. Georgia (1972) The arbitrary and inconsistent imposition of the death penalty violates the Eighth and Fourteenth Amendments, and constitutes cruel and unusual punishment. led to a de facto moratorium on capital punishment throughout the United States. In order to reinstate the death penalty, states had to at least remove arbitrary and discriminatory effects Gregg v. Georgia (1975) Led to USSC narrowing range of death-eligible crimes No more mandatory capital punishment for designated crimes Guided discretion Aggravating and mitigating circumstances Standard of proof higher for aggravating factors (beyond a reasonable doubt), and everyone has to agree on these Proof is lower for mitigating, and not everyone has to agree on which mitigating factor is important, just that each person determines there is enough mitigating circumstances Estimating incidences of wrongful capital convictions and executions Gross & O'Brien (2014) 1 in 25 capital convictions between 1973 and 2004 were wrongfully convicted Rate of innocence must be higher in LWOP - neither executed nor exonerated, but forgotten 30 states currently authorize capital punishment Highest rate per year was in 1999 (98), but has been steadily falling The Constitution and risk of wrongful executions Kansas v. Marsh If the law that "If a unanimous jury finds aggravating circumstances are not outweighed by mitigating circumstances, the death penalty shall be imposed" is constitutional or not "Courts do not find people guilty or innocent... a not guilty verdict expresses no view as to a defendant's innocence. Rather, a reversal of conviction indicates simply that the prosecution has failed to meet its burden of proof" (p. 647) Utterly impossible to regard exoneration as a failure of justice system, rather than a vindication of its effectiveness in releasing not only defendants that are innocent, but those whose guilt has not been established beyond a reasonable doubt. Final decision: Kansas law is unconstitutional because it mandates death based on a pro and con list Baze v. Rees (2008) Argued that the lethal chemicals (specific "cocktail" used in all lethal injections" used carried an unnecessary risk of inflicting pain during the execution Three purposes of death as a sanction: incapacitation, deterrence, and retribution - all have been called into question Final decision: The Supreme Court upheld Kentucky's method of lethal injection as constitutional by a vote of 7-2. No single opinion carried a majority. Glossip v. Gross (2015) 20 condemned prisoners petitioned for a writ of certiorari and stays of their executions from the U.S. Supreme Court. The petitioners argued that the midazolam, intended to be used as sedative, would not render them unable to feel the pain of the other two drugs. Therefore legal injections constitute cruel and unusual punishment under the Eighth Amendment. Today's administration of death penalty involves three fundamental constitutional defects serious unreliability = cruel Arbitrariness in application = Unfair long delays that undermine penological purpose = cruel Final decision: the Court held that lethal injections using midazolam to kill prisoners convicted of capital crimes do not constitute cruel and unusual punishment under the Eighth Amendment to the United States Constitution. The Court found that condemned prisoners can only challenge their method of execution after providing a known and available alternative method. Legislative reforms to reduce risk of error No system could ever be constructed that would perfectly guarantee that absolutely no innocent person is ever again sentenced to death Massachsetts suggestions (1) a narrowly defined list of death-eligible murders; (2) appropriate controls over prosecutorial discretion in potentially capital cases; (3) a system to ensure high-quality defense representation in potentially capital cases; (4) new trial procedures to avoid the problems caused by the use of the same jury for both stages of a bifurcated capital trial; (5) special jury instructions concerning the use of human evidence to establish the defendant's guilt; (6) a requirement of scientific evidence to corroborate the defendant's guilt; (7) a heightened burden of proof to enhance the accuracy of jury decision making; (8) independent scientific review of the collection, analysis, and presentation of scientific evidence; (9) broad authority for trial and appellate courts to set aside wrongful death Sentences: (10) the creation of a death-penalty review commission to review claims of substantive error and study the causes of such error Maryland Reforms A defendant found guilty of murder in the first degree may be sentenced to death only if: . . . the State presents the court or jury with: biological evidence or DNA evidence that links the defendant to the act of murder; a video taped, voluntary interrogation and confession of the defendant to the murder; or a video recording that conclusively links the defendant to the murder. . . . Death-Qualified Juries Witherspoon v. Illinois (1968) Death-qualification by 100% automatically one way or the other) at sentencing phase is fair, but unfair at guilt phase Wainwright v. Witt (1985) Overruled Witherspoon Excusal for cause of venirepersons whose opposition to the death penalty would prevent or substantially impair the performance of their duties as jurors during either: (1) the guilt phase of the trial; or (2) the sentencing phase of the trial. . . . Lockhart v. McCree (1986) Murder defendant's claim that the death-qualification process violated his rights to a trial by a representative and impartial jury. Argued that Death-qualification of his jury resulted in a "conviction prone" panel Tried to show a relationship between death penalty attitudes and verdict preference with 14 social science studies; that these juries are unrepresentative and predisposed to a particular verdict USSC dismissed each study, claiming invalid - but even if it was valid, it's immaterial to the law "Juries are made up of impartial jurors who met the standard of juror eligibility" But literally all evidence points that jurors are at risk for being conviction prone Morgan v. Illinois (1992) Must be willing to consider death (Death-Qualified) Must also be willing to consider life to be life (Life-Qualified) Current Rule (2019) Witt AND Morgan State v. Griffin (1999) The defendant appeals from the judgment of conviction. In her appeal, the defendant claims that: (1) the trial court improperly permitted the state to "death qualify" 2 potential jurors prior to the guilt phase of her trial, thereby denying her right, under the Connecticut constitution, to trial by an impartial jury; (2) the trial court improperly excused a venireperson; and (3) the trial court's instruction on proximate causation was flawed. Social science studies (14) showed death qualifications = conviction prone jury But small problem with every study (attitudes not actions, not in real life under oath, etc.) so court threw them all out Would anything be good enough?? Capital felony trials should have the same jury serve at guilt and penalty phases Efficiency Same information and evidence (aggravating factors) in guilt phase as sentencing phase, same jury takes less time Final decision: Defendant's claim fails. The judgment is affirmed, and the appeal denied.

Maeder and Pica, 2014

INFORMANTS Abstract: The goal of this research was to determine whether the size of the incentive (none, small, medium, or large, in terms of sentence reduction) a jailhouse informant receives for testifying, as well as scientific expert testimony regarding the fundamental attribution error, would influence mock juror decision-making in a criminal trial involving a secondary confession. Participants read a murder trial transcript involving informant testimony in which incentive size and expert testimony were manipulated and then provided verdict judgments, made attributions for the informant's decision to testify, and rated the informant and expert on a number of dimensions. Neither expert testimony nor size of incentive had a direct influence on verdicts. However, contrary to previous research on the influence of incentives on jurors' perceptions of secondary confessions, the presence of an incentive did influence verdict decisions, informant ratings, and attributional responses. Results imply that jury-eligible community members may be becoming aware of the issues with informant testimony as a function of incentive but that they are insensitive to the size of the incentive, and expert testimony may not sensitize them to the limitations of such testimony.

Key et al., 2018

INFORMANTS Abstract: We surveyed students, community members, and defense attorneys regarding beliefs about secondary confession evidence (i.e. when a third party tells authorities that a person has confessed to him or her) from jailhouse informants and other sources. Results indicated that laypeople perceive secondary confessions as less credible than other types of evidence (e.g. forensics, DNA, eyewitness testimony), and they are knowledgeable about factors that may influence the veracity of secondary confessions, such as incentives or previous testimony. However, they underestimated or were uncertain about how persuasive secondary confessions would be to themselves or other jurors. Compared to laypeople, defense attorneys were more sensitive about issues affecting the reliability of secondary confessions.

Cohn-Bucolo and Sommers, 2012

RACIAL ISSUES As of 2010, 60% of exonerees involved a black defendant Cross-race bias (AKA other-race effect) Eyewitness ID This chapter focuses on differences in how the legal system treats Whites and Blacks Study of racism and prejudice history Abnormal B Could be expressed by ordinary citizens Development and maintenance occurs naturally through social and cognitive processes Encourages through societal structure Current research Implicit vs. explicit prejudice Theory of Aversive Racism Socialized that racism and discrimination are wrong, so openly report explicit attitudes that align with this BUT society still reinforces and maintains negative stereotypes Implicit biases are formed, which can unknowingly affect B Racism seeps out in subtle and discreet ways where actions will not seem influenced by race More likely to be exhibited in situations where whites B can be rationalized and blame other factors and not race METHODS Secondary Analyses Reexamine existing data sets From one study Or meta-analyses Downside No control over methodological issues Dependent measures, Sample size, etc Public Opinion Polls Often national polls Telephone surveys, online, etc Downside Self-report Mock Juror lab studies College students or more representative samples Manipulate Opening/closing statements Testimony of defendant Jury instructions Diversity of the jury Self-report perception and B studies Ask ppts in legal system about perceptions and experiences Judges, attorneys, prosecutors, other employees RELEVANT RESEARCH Secondary Analysis of Arrests and Sentencing Racism towards Blacks occurs at many points in legal system Aversive Racism Drug use/crime example Prison sentence lengths are longer for Black MALE defendants One study found 116 differences in sentencing between Whites and Blacks Crime dependent Strongest effect in non-federal drug crimes No effect in non-federal property crimes Effects have been decreasing over time Largest disparities seen prior to 1970; smallest seen post 1990 Blacks more likely to receive death penalty in capital cases As of 2006 42% of death row inmates are black "race has the clear potential to influence police enforcement priorities, arrest decisions, and judicial and capital sentencing" Racism in Decision Made by Mock Jurors Effect of defendant race in mock juror research hasn't always been consistent Meta-analyses find different results as well In theory due to changing nature of racism toward blacks in modern legal system Aversive racism approach now When jury instructions include statements to avoid prejudice or biases, jurors tend to be more sensitive towards race Any racial framing will cause this to happen; interracial nature of crime, focusing on how it was a racially-charged crime, etc When racial issues made salient, racial bias decreased in both prejudiced and non-prejudiced individuals Racial Profiling and Treatment by Police Either focusing on objective B perspective or a subjective perceptual perspective Objective B Baseline/Benchmark Research Recorded race of drivers then race of drivers stopped - then compared percentage of minority drivers with percentage of minority drivers stopped = black drivers disproportionally likely to be pulled over Subjective Perception Public Perception Research National or local public opinion polls of citizens' perceptions and experiences Typically on how widespread racial profiling is, whether it's justified, and whether they feel like they've been stopped b/c of their race or ethnicity Blacks respond as expected Inconsistencies exist when Hispanics measured Police Officer's Perception One very preliminary study (Glover, 2007) Acknowledged racial profiling but rationalized use MOVING FORWARD: APPLICATION OF RESEARCH Racism currently is subtle and complex Research needs a multiple systems approach specifically designed to curtail the different racial biases that exist Both national programs and local communities needs to establish policies Based on literature Eliminate racial discrepancies Reduce minority disadvantages Policies need to be reformative Proactive Based in aversive racism theory Accountability is key - norms predict B Common In-Group Identity Model (Gaertner & Dovidio, 2000) Aimed at reducing prejudice among competing groups and reducing out-group bias through cooperation Through group recategorization Share membership in some "superordinate identity" as well as in their original groups Positive contact between outgroup members reduces bias Must be purposeful with the intention to reduce bias towards minorities Could be applied to police officers and court officials Positive contact and interactions are crucial to move forward Could eliminate the "us" vs "them" mentality Effect has been seen in Trial Simulation Research Interracial juries reduced racially biased decisions against defendant

Meissner et al., 2005

RACIAL ISSUES Cross-race effect (CRE) AKA own-race bias, other-race effect Faces of own race are better remembered when compared with performance on faces of other races Potential Social-Cognitive Mechanisms Ingroup/outgroup model (IOM) of the CRE Interracial Contact Reducing stereotypic responses and increasing the likelihood to look for more individuating information Influencing motivation to accurately recognize other-race persons through associated social rewards/punishments Reducing the perceived complexity of other-race faces Cognitive disregard and categorization effects People will cognitively disregard or render invisible others based on categories such as age, sex, or attractiveness Rely on this to determine in further interaction should occur Could apply to race to this Out-group homogeneity effect Outgroup members are often perceived to be more homogeneous and stereotypic than ingroup members Race-feature perspective Automatic inclination to racially stereotype/categorization may distract from encoding of individuating information Level of processing Memory strength depends on the depth to which it is processed during encoding 1981 study found Own-race faces processed at deeper level BUT no one has been able to replicate Distinctiveness Effects Novel faces that are distinctive or unusual are recognized better Using the Multi-Dimensional Space (MDS) Framework, examined effects of race, typicality, and level of perceptual experience within cross-race paradigm Distinctiveness for low-contact faces were confined to own-race High-contact faces were rated as highly distinctive regardless of race Configural-featural hypothesis Face inversion effect found to be not unique to faces, but highly encoded stimuli Expertise leads to configural (relational) processing One study found that own race faces were more susceptible to inversion than other-race faces both in reaction time and accuracy BUT other studies have not been able to replicate Ingroup/outgroup model Attempted to bring together all of the above A default or automatic process occurs when encounter own-race face Deep encoding Focus on relevant, configural properties Racial characteristics of other-race face signal an automatic categorization response Cue to cognitively disregard faces Attentional processes may be allocated elsewhere May lead to shallow/featural based encoding May signal stereotyping processes Dual Process Recognition Recall Episodic info is retrieved at time of recognition 'Remember' judgements Fluency General familiarity is used as basis 'know' judgements Application to CRE No studies have attempted yet, but IN THEORY Racial encoding: greater influence of familiarity Encoding based Processes: greater influence of recollection Configural-featural processing: greater influence of familiarity Experiment 1 Students rated both black and white faces for perceived distinctiveness, familiarity, likability, attractiveness, and memorability Findings Own-race faces were perceived as having greater memorability and familiarity Superior encoding leads to increases in both recollection and familiarity based responding Experiment 2 Tested dual-process predictions of the CRE by assessing the contributions of recollection and familiarity to performance on own and other race faces in a recognition memory paradigm First: asked participants to provide phenomenological judgements of recognition using Remember, Know, and Guess alternatives Second: using a stimulus based approach, perceptual ratings of memorability and familiarity from Exp 1 were used to predict recognition performance on own and other race faces in Exp 2 Findings Memory for own race faces involved greater recollection of info from the prior study when compared with performance of other race faces Demonstrated a differential effect of memorability on recognition of own vs other race faces Memorability significantly influenced both hit and false alarm responses to own race faces, whereas familiarity failed to differ as a function of race of face General Discussion The CRE appears to be due to a greater reliance on recollection for own-race faces Role of recollection is consistent with the differential effects of distinctiveness in CRE BUT Recollection in CRE does not provide support for reliance upon configural vs featural information or for racial profiling/stereotyping The CRE occurs b/c people encode more qualitative information about own-race faces, information that is diagnostic for subsequent identification situations Current study demonstrates that more incorrect identifications are made with high confidence for other-race faces BUT there is no measure at present that would be forensically useful in predicting which individuals are most likely to manifest a strong CRE in face identification

Discovery-From CrimePro

TUNNEL VISION Criminal Discovery Rules Disclosure duties on prosecutors Perjured testimony Material evidence favorable to accused (Brady issues) Rauland Grube v. State (2000) Grube found guilty of murder Gifford testimony never disclosed Information about Grube's reaction to news about death Implicated other suspect (police officer Brood) by questioning alibi Contradicted state's primary witness All referred to as Brady Material Tunnel vision in Grube Case Withheld evidence Doctored police logs Evidence appearing after years Expert eliminating Grube's shotgun as murder weapon Did not try to prove Grube had a vehicle when the killer did Fingerprint did not match Brady v. Maryland Rules Prosecution is bound to disclose all exculpatory evidence, including impeached evidence Reviewing court must determine if omitted evidence creates a reasonable doubt that otherwise did not exist If verdict is already of questionable validity, additional evidence of relatively minor importance might be sufficient to create a reasonable doubt Not about different verdict, but about receiving a fair trial Three components of Brady violation Evidence must be favorable to accused, either because exculpatory or impeaching Evidence suppressed by state, either willingly or inadvertently Prejudice must have ensued About fairness of trial Notes on Brady Evidence favorable to the accused: majority position Applies in every criminal case, though rarely used Exculpatory or impeachment of evidence in government hands Prosecutor does not have to know about evidence, but still can be blamed for not knowing - they have a duty to inquire Materiality of exculpatory evidence Defendant must show a reasonable probability that verdict would have been different Most courts assume burden lies on defendant, but no official rule Brady and plea bargaining No official rule, but some states link disclosure obligation to a defendant's not-guilty plea at arraignment Access to confidential information The judge under federal law can make the discovery ruling without an adversarial hearing Preservation of evidence States duty to preserve evidence only if evidence might be expected to play a significant role in the suspect's case In some states, trial judge determines breach of government duty by Degree of negligence or bad faith involved Importance of missing evidence in light of remaining evidence Sufficiency of other evidence to sustain the conviction Disclosing false testimony Includes deals or arrangements made for testimony Whether there is a reasonable probability that false evidence may have affected the judgement of the jury Extent of disclosure violations Difficult to document and rarely lead to reversal of a conviction Remedies for prosecutorial disclosure violations State bar authorities discipline criminal prosecutors less than they discipline private attorneys Discipline rarely occurs after first offense at state criminal level

Acker and Redlich, 2011

WRONGFUL CONVICTIONS OVERVIEW a. APPROACH i. Overview chapter on Wrongful Convictions b. MAIN FINDINGS i. Definition: "Factually innocent people are convicted of crimes" ii. Ways to measure wrongful convictions 1. DNA exonerations 2. Real perp confesses 3. Eyewitness recants 4. New evidence iii. Guilty pleas 1. Lesser charge/shorter sentence 2. Protect others 3. Can't afford bail, get out with "time served" 4. Prosecutors make mandatory minimums at court threat 5. BUT waives right to appeal c. IMPORTANT TAKEAWAYS i. Hard to fix a system that doesn't think it's broken

Feld, 2014

AGE/VULNERABILITY APPROACH Chapter on adolescent competence and the juvenile court system MAIN FINDINGS Courts use adult standards Competence to stand trial Waiving Miranda Waiving right to counsel Exceptions Roper v. Simmons (2005) Nationally cannot sentence death penalty to juveniles Graham v. Florida (2010) Nationally cannot sentence LWOP to juveniles (for non-homicide offenses) Adolescent Development and Juvenile Justice (ADJJ) Studied decision making, judgment, competence, culpability Children DO NOT EQUAL Adults Found that immaturity level of children is equal to the exclusion standards for incompetency IMPORTANT TAKEAWAYS No counsel Parents are allowed to be present during interrogation, sentencing, and intake - but not legal counsel Parents DO NOT EQUAL counsel Nor do school resource officers or probation officers Diversion is basically a plea bargain, but no legal counsel allowed Counsel only appointed if "severe" sentencing is expected No jury Most states won't allow juveniles a jury trial Reasoning: juries would ruin the informal and flexible nature of juvenile court, making it the same as adult court Yet still have a suppression hearing Judge has access to all case materials Too much power in one person No appeals No counsel Not long enough sentences (on average) to appeal Process isn't well documented Informal Little to no proof or evidence for appeal anyways Needed Reforms A mandatory and non-waivable appointment of counsel The right to a jury trial

Schreiber et al., 2006

AGE/VULNERABILITY APPROACH Compared real world data (McMartin Case, Michaels Case, and Child Protective service interviews) for suggestive procedures METHOD McMartin Case Seven teachers accused of flying students out to farm, saw animal torture and forced orgies Michaels Case Teacher raped them with various objects, forced them to eat her poo and pee, and to make a naked pentagram with their bodies Control/Child Protective Service (CPS) Interviews Not a perfect control group, but for a real-world data study, pretty good Measures Length of interview Total length Interviewer # words spoken Child # words spoken Ratio I words to C words Question Format Open/Narrative Yes/No Choice Focused/Specific (who what when where) Suggestive Techniques Reinforcement Repetition Other people (co-witness) info Speculation New Information MAIN FINDINGS Both cases Suggestive techniques Reinforcement Co-witness info New info McMartin only Suggestive techniques Speculation/pretend Use of puppets "Fun" Positive reinforcement Question styles Yes/No Focused/Specific (who what when where) Length 3x longer than both Michaels and CPS Higher interview ratio for words (most of the talking) Michaels only Suggestive techniques Negative reinforcement Doubting the child is telling the truth IMPORTANT TAKEAWAYS Four Red Flags to watch for Suggestive techniques Other People Speculation Positive Reinforcement Should only occur when building rapport New Information About perp or circumstances of abuse Negative Feedback NEVER express doubt about truth of child's statement Current analyses showed how social influences can impact real world situations Specifically how these influences are used with interviewing children Multiple ways to lead and/or contaminate an interview Various social influences can elicit same outcome Created methods for future analyses FLAWS Not representative of all interviewing techniques Only focused on one source of social influence - interviewers' questions Relied on interview transcripts, did not use audiotapes or videotapes Control may not have been representative of general practices Ages of children in control were slightly higher than McMartin or Michaels

Owen-Kostelnik, Reppucci and Meyer, 2006

AGE/VULNERABILITY Testimonial Evidence from Young Witnesses Taint hearings to assess suggestibility of child witnesses Incapacity below a specified age, and those requiring an understanding of the oath involved Confession Evidence from Young Suspects Youth is a risk factor for providing a false confession 12-13 year-olds most likely to confess in Alt Key cheating paradigm In re Gault and In re Winship - giving constitutional due process privileges against self-incrimination and beyond a reasonable doubt standard of proof Right to an attorney? ... if parent pays for attorney Assessing comprehension of miranda rights is not required, only some officers test comprehension Reid technique is employed, children more likely to falsely confess Review of Developmental Research Suggestibility Effect of Age on observing and remembering Effect of Age on Verbal Abilities involved in communication Effects of authority figures Immaturity of Judgement Immaturity Within the Procedural Contexts of Interrogation Miranda Rights Interrogation Confession The rest of this talks about cases and social reforms

Kassin et al., 2010

AKA "THE WHITE PAPER ON POLICE INDUCED CONFESSION" APPROACH White paper on Police Induced False Confession. Identifies suspect characteristics, interrogation tactics, and the phenomenology of innocence that influence confessions MAIN FINDINGS American interrogation: accusatorial approach Behavior based deception detection Psych methods of manipulation Maximization Guilt is certain, denying is useless Minimization Provide excuse/justification to make crime morally acceptable Use of False Evidence False Assertion vs False Physical Evidence Types of False Confessions Voluntary Fame, Self-punishment, Mental Illness, Protecting Others Compliant Get out of current situation, Avoid negative consequences, Reward (explicit or implicit) Internalized Come to believe they did it, can even create false memories Court's response Miranda Rights vs Warnings Nonspecific about language Not obvious to general public Waiving increases risk of false confession phenomenology of innocence: innocence is transparent, believe in the system Corroboration Doctrine Confessions backed up by evidence (details of crime that match confession) Voluntariness Reliable 5th Amendment (against self-incrimination) 14 Amendment (Due Process) Totality approach Risk Factors Situational (system variables) Isolation and Custody Use of false evidence Minimization (theme development in Reid) Dispositional (estimator variables) Age and Maturity Cognitive and Intellectual Impairments Personality traits and Mental illness Innocence Innocence is transparent DNA/other evidence will prevail Don't need Miranda BUT Reid encourages behavioral confirmation Police have prior belief about a suspect, treat suspect as guilty - innocent suspect firmly denies guilt, which then is taken to be suspicious and guilty behavior IMPORTANT TAKEAWAYS American interrogation system built to be coercive and deceptive Situational risk factors Innocence as risk factor Court's protections against coercion usually aren't successful Dispositional risk factors Best solution for now Record all interrogations from start to finish

Olson and Wells, 2004

ALIBIS Abstract: A taxonomy of alibis is proposed involving two forms of supporting proof: physical evidence and person evidence. Levels of physical evidence and person evidence were combined to create 12 cells in the taxonomy. Participants (n = 252), who were asked to assume the role of detectives, evaluated alibis representing these 12 cells. The believability of the alibis generally followed the taxonomy's predicted pattern, but physical evidence, when present, tended to overwhelm the person evidence more than had been expected. In addition, alibi evaluators seemed to not consider the possibility that a stranger who corroborated an alibi might be mistaken about the identity of the person. Trait inferences regarding the alibi providers tended to follow the believability data, even when the traits themselves were not relevant to believability of the alibi. We call for the development of a literature on the psychology of alibis, recommend the taxonomy as a framework, and suggest several avenues of inquiry.

Charmen et al. chapter in press

ALIBIS Discusses alibi generation, believability, and future directions Previous research on estimator variables; for real world, need to shift to system variables Alibi generation Guilty suspects Unverifiable details Choose easily manipulated corroborators Rarely provide false physical evidence Generalization hard to do Innocent suspects Autobiographical memory recall Errors of omission and commission occur Main obstacles A lack of memory of one's whereabouts - but a willingness to report an alibi nonetheless Overreliance on schema-based responding Difficulty of providing corroborating evidence The schema disconfirmation model Memory search Memory assessment Another memory search with schema as cue Either don't find disconfirming evidence Or do May act as cue to trigger new memory searches Alibi believability Alibi itself Supportive evidence Consistency of alibi Salaciousness of the alibi Watching porn can sometimes increase believability External factors Evaluation context Presence of other evidence Defendant characteristics Alibi skepticism hypothesis "alibi" may be loaded term New proposal Skepticism arises due to high expectations Skepticism arises when have to reconciling with contradictory evidence Future directions Further develop our theoretical understanding of the alibi process Develop interventions to differentiate the types of alibis provided by innocent and guilty suspects Develop new paradigms to assess alibi accuracy Increase the diversity of participant samples

Leins and Charman, 2016

ALIBIS Goal was to develop an objective alibi paradigm Tested time and location cues on accuracy of generated alibis, as well as how schema consistency impacted accuracy and confidence ratings Time cues Ineffective because people don't auto-encode time Probe 3 levels of memory Abstract conceptual knowledge Schematics of repeated events (what do I normally do during that time) Episodic memories Often stops at schematics Accuracy Events must be encoded Retrieval cues need to activate those details Experiment 1 Tested time, location, and time/location Accuracy low in all conditions Location was higher than other two Schemas retrieved faster because they are better encoded than time? Efficiency/specificity tradeoff? Time cueing increases schema use and decreases accuracy rates when event is not consistent with schema Experiment 2 Manipulated schema consistency for accuracy; also measured confidence (metacognitive monitoring) Schema consistent, increased accuracy and confidence Alibi reports based on schemas Did not control for how memorable event was though Confidence high in all conditions Schemas easily accessible - mistake familiarity with accuracy? Schemas naturally contain plausible locations Overall discussion Was able to create objective alibi paradigm Alibi providers use schemas, which increase risk of confidence in inaccurate alibis May be due to overconfidence as a result of ineffective cueing General accuracy of schema use depends on if event was schema consistent Most people provided an alibi Believe innocence transparent Retrieving a schema is easy Easy = accurate

Dysart and Strange, 2016

ALIBIS Notes: 25% of 157 exonerees provided an alibi 100% of them had one Goal of current study: interview police officers about alibi beliefs and practices Olson and Wells (2004) 2 forms of alibi evidence Physical evidence (none, easy to fabricate, difficult to fabricate) Person evidence (none, motivated familiar other, non-motivated familiar other, and non-motivated stranger) Evaluation phase Point where credibility is determined matters During investigation or in court Mistaken Alibis Olson and Charman (in press) 36% of people gave mistaken alibi Memory for details of events fade Higher rates of encoding for unusual events, therefore higher recall Prior Notice Rule Defense must present alibi to prosecution before trial or could be excluded Results 2 themes emerged Time Recognized that the sooner acquired/vetted, the better Evidence less likely to be altered or disappear Memories stronger But current procedures increase delay time Person who receives alibi usually is not the one to investigate it Distrust of suspects 81% said if suspects change their alibi, they were lying originally Take together Police think suspects are lying, don't want to waste resources investigating a lie (low priority), increases delay, increases chances that corroborator will misremember or forget

Charman et al., 2017

ALIBIS Notes: Main questions/goals Can a potential alibi corroborator recognize a stranger that they have met? Can accuracy of corroborators be improved? Can alibi generators accuracy determine how likely it is that their corroborator remembers them? Is there anything else that correlates with corroborator accuracy? Limited research on alibi generation Common mistake to mistakenly give a wrong alibi Persuasive corroborators are most often strangers (not friends/family) Previous studies used eyewitness paradigm, not alibi Alibi paradigm Expose ½ corroborators to target, then show all corroborators a target-present ID task 3 reasons why corroborators are bad at facial recognition No reason to encode face No reason to encode context or ID of target Lengthy time delay Usually several days Can accuracy of corroborators be improved? Cueing time Increase correct IDs Using photo lineup over showup Decrease false IDs Can alibi generators accuracy determine how likely it is that their corroborator remembers them? Spotlight effect Everyone notices me Took target confidence ratings after interacting with corroborator Is there anything else that correlates with corroborator accuracy? Duration of interaction Retention interval General likelihood of encountering strangers Distinctiveness of target appearance Results 37% of corroborators accurately ID'd target But only 7% ID'd target, time of interaction, and nature of interaction Cueing didn't have an effect Small indication that showup increased accuracy Targets were highly overconfident that they would be remembered Only 5% false ID's Why different than eyewitness? Different encoding (mundane vs. salient)? Raised decision criteria? Less pressure to make an ID? Future research Look more into full context reinstatement (not just time) to see if improves accuracy

Drizin et al., 2014

APPROACH Chapter on stereotypes/teenage tropes, false confessions, and narrative contamination MAIN FINDINGS Narrative Contamination: contextual details that cannot be proven by evidence Make for an authentic, compelling, and self-corroborating narrative Preconceived notion of suspect or crime EX: car theft: white woman suspect or black male suspect? Going to choose black male Stereotypes Super Predator African American males who had bad childhoods Animalistic, travel in packs Examples Central Park Jogger Case 5 youths convicted of rape of white female jogger Dixmoor Five Case 5 youths convicted of rape/murder of 14 yr old girl Englewood Four Case 4 youths convicted of rape/murder of addict/sex worker Angry Teenage Loner White males who are ostracized by society, are bullied or picked on Violent video games/movies "gave them permission" Jealous, resentful Examples School Shootings (Columbine) Michael Crowe Younger sister stabbed Thomas Cogdell Younger sister suffocated IMPORTANT TAKEAWAYS 95% of confessions later exonerated by DNA included accurate detailed information ("known only to perpetrator") Juveniles more likely to falsely confess and are more prone to narrative contamination Solution: Record entirety of interrogations & confessions

Pimentel et al., 2015

Abstract: False confessions are often involved in wrongful convictions and are sometimes made to protect someone else (i.e., the guilty), perhaps as a way of reciprocating past favors. Experimental research has neglected to investigate false confessions made to protect someone else, including among adolescents who may be particularly vulnerable given their peer orientation and sensitivity to peer influence. The present study examined (a) how often individuals would falsely confess to take the blame for another's wrongdoing, and (b) whether the willingness to falsely confess would vary by age group and reciprocity. Adult (n = 99, M age = 20.29) and adolescent (n = 74, M age = 15.47) participants were randomly assigned to either receive or not receive a small favor from a confederate, witnessed the confederate cheat on a task, and then decided whether to sign a statement taking the blame for the confederate's cheating. Adolescents (59%) were more likely to falsely confess than adults (39%). No effect of reciprocity was found. Although a well-documented phenomenon in legal cases and field studies, the present study provides the first experimental evidence concerning voluntary false confessions to protect another, including developmental differences in this tendency. Results highlight the vulnerability of youth and the potential role of peers in legal decision making.

Russano et al., 2005

Abstract: The primary goal of the current study was to develop a novel experimental paradigm with which to study the influence of psychologically based interrogation techniques on the likelihood of true and false confessions. The paradigm involves guilty and innocent participants being accused of intentionally breaking an experimental rule, or ''cheating.'' In the first demonstration of this paradigm, we explored the influence of two common police interrogation tactics: minimization and an explicit offer of leniency, or a ''deal.'' Results indicated that guilty persons were more likely to confess than innocent persons, and that the use of minimization and the offer of a deal increased the rate of both true and false confessions. Police investigators are encouraged to avoid interrogation techniques that imply or directly promise leniency, as they appear to reduce the diagnostic value of any confession that is elicited.

Smith & Haney, 2011

DEATH PENALTY Every study finds that juror understanding of legal instructions is very low Specifically, even if you increase declarative knowledge (definitions) they still suck at procedural knowledge (applying the term to a scenario) Two phase system Guilt phase Sentencing phase Death or LWOP Research has shown that capital sentencing instructions are incomprehensible Nature of key terms used Courts aren't required to define anything Aggravation (aggravating factors) Support DP Mitigation (mitigating factors) Support LWOP Lack of any clear explanation provided Ambiguity in structure of decision making process How are they supposed to weigh factors? Study One Examined if new CA instructions increased comprehension of capital sentencing instructions Participants heard either standard CA instructions (CALJIC) or newer CALCRIM) read aloud to them twice, then asked comprehension questions RESULTS Test scores of CALCRIM (new) were significantly higher Definitions of key terms were significantly better in CALCRIM Comprehension of factors significantly better in CALCRIM No different in weighing process DISCUSSION Overall comprehension still low Improved declarative knowledge, but not procedural Better at identifying aggravating than mitigating factors Ability of participants to correctly apply key concepts to weighing scenario was low Death is never legally mandated, only situations where you can give it; however there are certain situations where you have to give LWOP This may have been confusing to participants when weighing Study Two Created and examined pinpoint instructions to improve juror comprehension Provide jurors thematic, case-specific examples of aggravating/mitigating factors in a relevant and concrete way Participants assigned to Patterned (CALJIC) vs. psycholinguistically improved vs. pinpoint Watched video of death penalty case (sentencing phase only) Previously convicted of first degree murder in the guilt phase, participants' task to give sentence RESULTS Test scores: pinpoint and psycholinguistic significantly higher than standard Definition of terms: Pinpoint sig. better, then psycholinguistic, then standard Comprehension factors: pinpoint and psycholinguistic significantly better than standard, but overall still low Weighing: Pinpoint better than others, psycholinguistic & standard both equally not good Verdict choice: did not vary by instructional format DISCUSSION Pinpoint better, but is it good enough? Limitations: Instructions included psycholinguistic aspects, were longer, and included more repetitions of key terms USSC (as of 2011) has not acknowledged problem of instruction incomprehensibility

Bond and DePaulo, 2006

DECEPTION DETECTION Historical characterizations of Deception Ekman: It's in the hands and feet (1969) Ekman: There's no foolproof signs, it's ambiguous (2001) Face-to-face deceptive interactions Craft verbal deception, bolster it, and suppress discrediting B Buller and Burgoon (1996) Self-presentational perspective (DePaulo etc al, 1996) Lying is everyday, but stakes the liars reputation New Framework: a double standard Based on truth as a virtue; truth at face value Two perspectives How liars regard themselves Nothing exceptional; easy to rationalize How liars regard other liars Immoral, deception is wrong Essentially, Fundamental attribution error Stereotype of liars Tormented, anxious, conscious-stricken Detection Accuracy Liars as "senders", the statements "messages", and the judges "receivers" Lie detection is about 57% The current review Summarizes evidence from 206 studies Assess accuracy of deception judgements by percentage correct Various situations Various factors on accuracy Medium of detection Liars motivation Judges expertise FINDINGS Lie detection accuracy: 54% KNOW THIS NUMBER FOR QUALS Statistically more Significant than chance Didn't matter situation Didn't matter how you measured success People are more inclined to judge messages as truthful Truth judgements: 56% Might be underestimated, because it's in context of "be on the lookout for liars" yet STILL had truth bias Lies are more detectable when they are heard than seen Motivated senders are more easily detectible when visually seen (no difference in vocal statements) REGARDLESS of liar or truth teller Unplanned lies were more easy to detectible Baseline exposure helps to distinguish, but still about at chance Still more likely to say truth Interaction did not make any difference STATISTICALLY NO SIGNIFICANCE BETWEEN EXPERTS AND NON-EXPERTS Double Standard Model Stereotype of liars rarely fits, because we use others' lie experience rather than our own Torment is only seen in motivation to be believed in high stakes situations It's about what liars LOOK like, not what they SOUND like Judges excuse lies of others to maintain exception for self Truth bias Calling people that you know are liars is wrong, mean, bad Baseline exposure Interaction

Mohan and Walker

EXPERT TESTIMONY APPROACH Law chapter analyzing how Expert Testimony is allowed MAIN FINDINGS Social Frameworks Fact perspective Repetitive Expensive Social Authority Perspective Treat Research like legal precedent Mix of Authority and Fact Testimony as written brief to judge to justify jury instructions Federal Rules of Evidence (FRE) - Rule 702 "If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise" 1st prong Peer reviewed, Testable, published, have error rate 2nd prong Must be relevant Frye v. United States (1923) "General acceptance" in scientific community only requirement Daubert v. Merrell Dow Pharma (1993) 702 supersedes Frye FEDERALLY Still state dependent Daubert: not about conclusions about sound methods First to call judges "gatekeepers" "General acceptance" isn't specified, BUT expert's information still needs to be reliable and relevant Research needs to be done prior to/in no connection with case GE v. Joiner Abuse of Discretion Standard Kumho Tire Company v. Carmichael All expert testimony falls under 702, not just scientific (technical etc. too) Judges are gatekeepers, not scientists Still required to perform gatekeeper duties R702 Amended in 2000 to reflect Daubert Judges as gatekeepers Applied to all testimony IMPORTANT TAKEAWAYS Jury instructions maybe not the best idea Experts aren't supposed to connect general research with specific case, but still often do Expert Testimony should NOT be allowed to connect general research to specific case Should be courts/jury's job

Levett and Kovera, 2008

EXPERT TESTIMONY Backrground Judges are gatekeepers Supposed to keep out "Junk Science" Not great at judging validity of evidence Procedural safeguards Cross-examination Presentation of contradictory evidence Judicial knowledge (Jury Instructions) Burden is then on the Jurors But jurors cannot evaluate validity well either All safeguards have been shown to be ineffective in helping jurors to discount flawed evidence Jurors Can Jurors Differentiate Between Valid and Invalid Scientific Evidence? Sample Sizes Chance/probability Construct validity Confounds Blind experiments Control groups Opposing Experts Can Opposing Experts Safeguard against unreliable Expert Evidence? Expert v. Opposing Expert on Juror Decisions Opposing experts can impact credibility of original expert, but no effect on decisions Expert v. Opposing Expert AND Valid v. Invalid Evidence ONLY TWO STUDIES Cutler and Penrod (1995) Opposing expert caused skepticism of ALL evidence presented Two experts going head to head = concepts behind evidence must be disputed within the field Devenport and Cutler (2004) No influence on decisions but original expert seen as less credible Current Study Heuristic-Systematic (HSM) & Elaboration Likelihood (ELM) Central v. Peripheral Processing Process new information Quality of fact v. Heuristics or other factors Sensitization or Skepticism? Having opposing expert teach about the science might SENSITIZE them to quality of research Could increase SKEPTICISM of all evidence by acting as a heuristic/cue for discounting evidence Can we use opposing expert testimony as a safeguard to teach jurors about the validity of evidence as presented in the initial testimony? DESIGN Jurors read written summary of child abuse trial Validity of Defense Expert Testimony (Valid, no control, no counterbalance) Presence and Content of Opposing Expert Testimony (none, standard [based on published criticisms], methodology [based on internal validity]) Rendered verdict Guilty or Not Guilty Rated experts and expert testimony Experts Credibility Trustworthiness Expert Testimony Quality of Defense Quality of Prosecution (opposing) RESULTS Main effect of study validity on defense credibility and trustworthiness in methodology conditions Defense Expert more trustworthy and credible in the valid evidence condition when o. expert presented methodology Significant effect of opposing expert on ratings of defense expert Defense Expert more credible and trustworthy and presented higher quality research when no opposing expert was present No interaction between opposing expert presence and study validity Only the opposing expert manipulation significantly predicted verdict Presence of opposing experts increased guilty verdict by 83% DISCUSSION Skepticism effect supported Sensitization effect not supported No evidence of opposing experts being able to safeguard against flawed scientific testimony Presence of opposing expert impacted jurors' verdicts and ratings of the defense expert, regardless of the content of the defense expert's testimony LIMITATIONS No expert testimony absent condition Not able to draw conclusions about expert testimony on verdict No court appointed expert condition Adversarial expert may be seen as hired gun to disagree with other side

Evidence Chapter, 2012

EXPERT TESTIMONY Looking at SOE (Strength of Evidence) Eyewitness Characteristics Age Competence and Trustworthiness = credibility Confidence Interaction with ID alone Consistency Affects credibility, unsure on judgements and verdicts Expert Witness Benevolent educator? Hired gun? Superfluous pontificator? Expert Witness Characteristics Unopposed v. opposed Whoever testifies first has more influence (story model) Content specificity and form Specific testimony (applied to case at hand) and clinical form (looking at holistic assessment, not risk factors) more influential Content domain Eyewitness Reliability Sensitivity v. skepticism Battered Woman Syndrome Medium influence, but depends on other factors Personal attributes Credentials Mixed results Pay Mixed results Popularity as EW Mixed results Gender No difference, unless gender sensitive case Type of Evidence Scientific Polygraph Little impact Blood and Hair Jurors misunderstand stats, hard to tell Prosecutor's fallacy Defense attorney's fallacy DNA RMP v. LE Probabilities = no effect, format = effect Errors v. expectancies Exemplar Cueing Theory Easier to discount DNA match if coincidental match imaginable (availability heuristic) Frame Likelihood of a match Form Frequency, probability, odds Size of referent population VERY POWERFUL on verdict Most influential predictor Other than fingerprints? CSI Effect Not on verdict Confession Evidence Demonstrative Visual Evidence Gruesome Images More influential the more emotional the crime Computer Simulation Might work better than demonstrative exhibits? Neuroimaging Tests Too early to tell; not very influential yet Conclusions Strong Evidence = Good Story Expert Witness Connections Most influential when: Apply knowledge to specific facts of case Provide new information not previously known Use scientific evidence to address ID In understandable format Otherwise will rely on heuristics like expert characteristics

Fitzgerald and Price, 2015

EYEWITNESS ID APPROACH A meta-analysis of age differences in eyewitness identifications. METHOD Examined Age Differences and Response Bias between Children, Young Adults, and Older Adults Eyewitness ID Target Present vs Target Absent Lineup Type (simultaneous, sequential, showup, elimination) Social vs Cognitive processes in each age group Response bias Choosing Need to ID a Suspect MAIN FINDINGS Main effect of age: young adults vs children/older adults Young adults Higher correct ID of TP Lineups Higher correct reject of TA Lineups Children AND Older adults Higher rate of choosing overall Lineup type Barely any effect on discriminability between guilty and innocent suspects Didn't moderate effect of age effects IMPORTANT TAKEAWAYS Children are less likely to correctly ID culprit in TP lineup, despite previous research claiming "children can tell" Children and Older adults made fewer correct rejections than young adults Lineup type did not moderate age effects on identifications FLAWS No longitudinal studies Signal detection measures used were not designed for eyewitness ID paradigms Did not analyze witness confidence

Wells and Quinlivan, 2009

EYEWITNESS ID APPROACH Analyzed two-prong approach for evaluating suggestiveness in eyewitness IDs for trial METHOD Two-pronged system for eyewitness ID First prong: Per se Exclusion Rule Unnecessarily suggestive procedures occur, evidence can't be used at trial If you can't have avoided doing a showup vs a lineup etc. Second Prong: Reliability/Totality Approach If unnecessarily suggestive procedures occur, then Manson "Biggers" Criteria must be applied to measure if the ID was NEVERTHELESS reliable Manson "Biggers" Criteria Opportunity to view Attention How do you measure? Description Accuracy vs. consistency Time to ID Certainty (Confidence) MAIN FINDINGS Three main themes in literature that affect Manson "Biggers" Criteria View, Attention, Description all Retrospective self-report Subjective, all highly malleable Nature of relationship between criteria and EW ID accuracy Criteria themselves not independent of each other Suggestiveness Augmentation Effect "Irony of Manson" - Suggestiveness influences self report, which can dismiss suggestiveness Does not deter suggestiveness; rewards it "Nevertheless" tendency of courts Can ignore one criteria in favor of another stronger one IMPORTANT TAKEAWAYS Manson/Biggers criteria stand in the way of eliminating suggestive EW ID procedures Future goals Alternative to Manson/Biggers Criteria Contain incentive not to use suggestive procedures Suggestive procedures confound entire process - need to be more thoroughly monitored Criteria need to be independent of suggestive procedure itself Shift the burden of reliability of evidence to prosecution Limit EW testimony Jury Instructions about suggestive procedures' impact on reliability

Greathouse and Kovera, 2009

EYEWITNESS ID Abstract: Pairs (N = 234) of witnesses and lineup administrators completed an identification task in which administrator knowledge, lineup presentation, instruction bias, and target presence were manipulated. Administrator knowledge had the greatest effect on identifications of the suspect for simultaneous photospreads paired with biased instructions, with single-blind administrations increasing identifications of the suspect. When biased instructions were given, single-blind administrations produced fewer foil identifications than double-blind administrations. Administrators exhibited a greater proportion of biasing behaviors during single-blind administrations than during double-blind administrations. The diagnosticity of identifications of the suspect in double-blind administrations was double their diagnosticity in single-blind administrations. These results suggest that when biasing factors are present to increase a witness's propensity to guess, single-blind administrator behavior influences witnesses to identify the suspect.

Kassin, 2005

FALSE CONFESSIONS APPROACH Review of the psychology of confession evidence and amplify the finding that "Innocence may put innocent people at risk" for false confession MAIN FINDINGS Types of False Confessions Voluntary Coerced-Compliant Coerced-Internalized Three main areas innocence may impact Pre-interrogation interview Waiving Miranda 4/5 suspects waive rights Kids, mentally impaired adults, and naïve adults lack comprehension Interrogation Reid (Isolation, Confrontation, Minimization) Encourages behavioral confirmation Only limitations No physical force or deprivation Threats of injury or punishment Promises of sentence leniency or full immunity Biggest issues Custody/Interrogation length Use of false evidence Use of minimization Issues Preventing Reform People trust confessions People suck at deception detection Police-induced confessions have all the elements of a true confession Details, emotions, etc IMPORTANT TAKEAWAYS Confession evidence is compelling Innocent suspects are more likely to waive Miranda rights (especially at-risk populations) Reid is confession-oriented, uses coercive and deceptive tactics, and does not contain protection for innocent suspects

Kassin et al., 2013

FORENSIC EVIDENCE Confirmation Bias and Forensic Evidence Forensic science not infallible Criticisms Can't be objective, so reliability based on subjective judgements Psychological biases Contextual factors Lack of objective standards "Sufficiently similar" Experts overstate strength of evidence Overconfident Courts blindly accept science Ground truth unknown Work for police/prosecution Consider themselves objective and immune to bias Cognitive Psych Expectancy effects Primacy effects Confirmation bias Visual Auditory Behavioral Self-fulfilling prophecy Heuristics Motivational goals Accuracy goals Directional goals Forensic Confirmation Bias "Class of effects through which an individual's preexisting beliefs, expectations, motives, and situational context influence the collection, perception, and interpretation of evidence during the course of a criminal case." (p. 45) Corroboration inflation Bias snowball effect Types of influential evidence (need ambiguity) Confession Handwriting analysis Polygraph results Fingerprint judgements Testimony of lay witnesses False eyewitness ID's Questioned Document Examination (QDE) Moderators can exist Ambiguity Eyewitness IDs Auditory IDs Polygraph results Big issue Many methods require matching visual evidence with sample from suspect Psychological contamination can occur at any point Reforms Reduce bias in lab Linear investigations Blind testing Evidence lineups Double blind verifications Candidate lists must be reasonable lengths and randomized Require basic psych understanding for relevant areas Reduce bias in courts Educate attorneys about analyst methods, practices, and basis for conclusions Inform judges and juries of potential bias/contamination Evaluate Harmless Error Doctrine decision case by case for potential corruption

Chapter 4 from Garrett, 2011

FORENSIC SCIENCE Forensic science cause of 74% of 250 wrongful convictions Most cases had analysts testify 61% of analyst testimony was invalid 81 different analysts 54 different labs/practices/hospitals 28 states Used to make forensic evidence stronger than it was Not neutral Testify for prosecutors, work for police Exposed to all facts/biases of cases Not well controlled; rarely audited or investigated Two Recurring problems Reliability Validity Methods Serology Reliable 67/116 cases = testimony invalid Error in masking Error in frequencies Microscopic Hair Comparison Unreliable No frequency stats No uniform # features to declare match 29/75 cases = testimony invalid 31 valid cases used misleading terminology Consistent, similar, match DNA Testing Reliable 16/20 cases DNA proved innocence, other 4 suggested potential guilt Only 8 confessed 5 cases even excluded defendant Invalid testimony Error in accuracy rate Error in statistics Error in lab Error in interpretation 'DNA is a truth machine that can be manipulated or abused' Bite Marks/Forensic Odontology Unreliable 5/7 cases = testimony invalid No case evidence of judge denying bite analysis American Board of Forensic Odontology Now doesn't allow certainty claims, but still happens Shoe Prints Reliable... ish? Match = narrowed down type of massed produced shoe/size Similar = might be helpful but unclear? Gait does not change shoeprints 1 case invalid testimony Scientific Working Group on Shoeprint and Tire Tread Evidence Voice Comparison Unreliable "an unreliable method that should have disappeared from our courtrooms decades ago" FBI won't allow it but some judges still will 1/20 cases = testimony invalid Lied about print Other types of error Concealing evidence Errors in analysis Failure to conduct elimination testing Lawyers and Judges 18 cases, prosecutor exaggerated science Judges usually do not deny evidence/conclusions; more likely to deny challenge of evidence Defense experts only in 21 cases Judges often deny funding for defense experts Reforms Independent crime labs Need quality control, independent oversight and auditing Judges keep out unreliable methods Lawyers question evidence, call for independent experts, keep analysts in check on stand (object) American Bar Association working on reforms Congress tasked NAS with improving quality They called for legislation

Semmler et al., 2012

JUROR KNOWLEDGE APPROACH Chapter on juror believability of eyewitness testimony Review on how courts expect jurors to evaluate EW Testimony, and the effectiveness of legal safeguards Review of experimental methods developed to assess how jurors determine accuracy of EW report and ID MAIN FINDINGS Biggers Criteria View, attention, detail, time, certainty All self report and vulnerable to feedback effects Three main categories of methodology to investigate juror beliefs in EW testimony Surveys of Jurors Usually find that laypeople and experts believe that confidence indicates accuracy, consistency of witness will predict reliability, and level of detail provided indicates accuracy Mock-witness or simulated trial methods Criticism Undergrads Less complex, less information given to participants Don't include effective cross-examination Lack stimulus variability (only one witness statement, doesn't combine evidence types, etc) Real-witness methods Wells, Lindsay, & Ferguson (1979) Three-phase paradigm Crime video, mock ID, mock juror assesses accuracy of ID Consistency Inconsistencies are viewed negatively Either lying or memory is bad, either way testimony is inaccurate BUT research shows that consistency of testimony does not mean every fact needs to be consistent Detail Tradeoff between number of details provided and accuracy Psych foundations Stereotypes Truth bias Testimony = memory, not method of questioning Fundamental attribution error Confidence heuristic Expert Testimony Three possible effects Confusion Increase sensitivity Increase skepticism Sensitivity vs Skepticism Jurors use signal detection theory to discriminate between target cue and noise Not all errors are created equal: adjust sensitivity, adjust how many errors you get IMPORTANT TAKEAWAYS Jurors use commonsense (EW's confidence) to determine accuracy Confidence heuristic Hold a truth bias Real-witness methods best external and internal validity

Mindthoff et al., 2018

JUROR KNOWLEDGE APPROACH Assess possible shifting beliefs in juror confession knowledge Online survey of potential jurors to evaluate perceptions of confession behaviors, Miranda waivers, interrogation methods, dispositional risk factors, and confession admissibility and weight METHOD 968 participants from 11 universities (biggest and most generalized sample yet in field) Amazon Mechanical Turk online survey Six topics of questions (with scales, sliders, and multiple choice used) General perceptions of confessions - (4 perceptions) Miranda waivers Perceptions of interrogation methods - (7 techniques) Relationship between dispositional risk factors and false confessions - (9 types) Admissibility of confessions and evidentiary weight in verdict decisions Personal characteristics (crime-media engagement and familiarity with disputed confession cases) - (4 types) MAIN FINDINGS Confessions Confessions as indicator of guilt 58% agreed confession=guilt Likelihood of false confession 62% agreed that suspects might confess to a crime they didn't commit Mental illness/physical torture 57% disagreed that these two are the only reasons people falsely confess Estimated occurrence rates People estimated that 30% of innocent people interrogated falsely confess Interrogations True evidence Believed highly likely to be used (true) Physical harm Believed least likely to be used (true) Rapport Believed sometimes used (Actually most used in real world) Dispositional Risks Believed all highly likely to contribute to false confessions Likely to be used: Mental illness (highest) Alcohol use (second highest) Low IQ (lowest) Under 18 (lowest) MJ use (lowest) Sleep deprived (lowest) IMPORTANT TAKEAWAYS Now 60% of jurors accept that suspects might falsely confess Jurors typically understand (more students than community members, but both) Suspects don't understand Miranda Police use manipulative tactics to get people to waive rights Innocent suspects more likely to waive rights Believe that coerced confessions shouldn't be allowed in courtroom Those that had been exposed to true crime media were more likely to perceive coercion More likely to suspect false confession FLAWS Possible that some participants were not jury eligible Generalizability of MTurk data Only assessed perceptions of general interrogation techniques categories, not tactics Results do not speak to juror sensitivity or skepticism regarding confession evidence

Beaudry et al., 2015

JUROR KNOWLEDGE APPROACH Experimental study using a real-witness method (three-phase paradigm [Wells et al 1979]) Looked at juror/observer perception/accuracy of eyewitness ID via evidence type, lineup presentation and lineup administration METHOD Participants watched an accurate or inaccurate mock ID video and/or a mock testimony Evidence type: ID or Testimony Lineup administration Sequential or simultaneous Admin knowledge (single blind or double blind) Admin feedback (present or not) MAIN FINDINGS Eyewitness evidence still more believable than not Truth bias Belief 5x higher when viewed testimony than ID decision Evidence type impacted response bias, but not sensitivity Also did not impact accuracy 53% accuracy rate for discriminating between accurate/inaccurate ID's Lineup presentation did not affect perceptions If exposed to ID video alone, simultaneous lineups actually improved accuracy Get rid of truth bias, increase skepticism Double blind vs single blind had no effect If witness received feedback, was seen as more believable Juror confidence in their judgements was not affected by accuracy of decision or accuracy of the eyewitness ID IMPORTANT TAKEAWAYS Video recording does not eliminate need for lineup administrators Also does not eliminate procedural biases Post-Identification feedback was most impactful on confidence FLAWS Did not view a complete court case with real attorneys and cross-examination Inaccurate ID condition included any selection from a TA lineup, instead of just choosing an innocent suspect EW and Admin were recorded in profile during ID, but EW was recorded head on during testimony

Palmer et al., 2016

JUROR KNOWLEDGE APPROACH First empirical test of whether inconsistencies undermine the credibility of evidence by manipulating consistency of confessions Background Inconsistency research has been focused on eyewitness or victim testimony Inconsistencies decrease credibility of witness - jurors more likely to discount testimony when making decision Several studies have examined the impact of retraction of testimony on verdict decisions If witness testimony, retraction accepted and testimony discounted If confession, retraction disregarded and included in verdict decision Other types of inconsistent testimony have only been examined anecdotally Anecdotal evidence states that inconsistencies have no effect on the perceived credibility of the confession or on the verdict METHOD Method: Experiment 1 Read verified informational sheet about crime Four versions: Consistent, Inconsistent, Coerced, Not Coerced Read signed confession Provide Verdict Survey Confidence in verdict, Probability of actual guilt, Consistency of evidence, Voluntariness of confession Method: Experiment 2 Read verified informational sheet about crime Four versions: Consistent, Inconsistent, Motive, No Motive Read signed confession Provide Verdict Survey Confidence in verdict, Probability of actual guilt, Consistency of evidence, Voluntariness of confession MAIN FINDINGS Findings - Experiment 1 Coercion and consistency: Significantly related In both coercion conditions, there were less guilty verdicts in the inconsistent condition Consistency, coercion and verdict: not sig. Related When coercion was present, inconsistent confessions had less impact on verdict than when there was no coercion present Findings - Experiment 2 Consistency and verdict: Significantly related More guilty verdicts in the consistent condition than the inconsistent condition Motive and verdict: Not significantly related The presence of alternative reasons for inconsistent confession did not impact jurors' verdicts Additional Exploratory Analysis When confession was inconsistent, participants were more likely to report that there must be an ulterior motive Not impacted by the Motive manipulation When there was an ulterior motive present, participants were less likely to perceive guilt Led to confession being discounted when giving verdict IMPORTANT TAKEAWAYS Inconsistencies CAN undermine credibility of confession evidence, despite previous anecdotal evidence stating that they don't Jurors less likely to give out guilty sentence when inconsistencies detected Inconsistencies lead jurors to either believe or generate alternative explanations for the confession FLAWS People could have seen a relationship between "other individual" and defendant Only looked at one type of inconsistency Used simple trial stimuli

Najdowshi, 2014

RACIAL ISSUES 47% of exonerees African American Focus shift in research Why are these innocents entering the justice system to begin with? Racial bias? Floyd v. City of New York 2013 90% of African Americans stopped and frisked by NYPD were innocent Chapter examines how cultural stereotypes of African Americans as criminals affect their reactions to police, and police judgements/treatment of them Focus on police encounters b/c this is first step in chain Stereotypes include Aggressiveness Tendency towards violence Criminality Hostility Defined by both races and those in high vs. low prejudice individuals Stereotype threat (when one acts differently because they are afraid to be judged unfairly) Stereotype might be causing AA to act more suspiciously around police, thus seemingly confirming stereotype Affects performance, anxiety Police often view reactions to stereotype threat as cues to deception Research findings Survey studies found AA more likely than whites to fell stereotype threat and worry/anxiety that is associated with it Psychologically real study confirmed main effect of race on nervous appearance mediated by stereotype threat Priming studies show link between crime and race Shooting studies Racial bias has been shown to influence shooting decisions Cultural stereotype can influence biases, even when not conscious to the perceiver Even if stereotype threat doesn't influence AA behavior, it may still influence police perception of that B (more deceptive and criminal) regardless Future Directions Research Need to study police bias expectations separately from AA bias expectations Also need to study behavioral confirmation - what happens AFTER initial contact Mistaken IDs an issue, even more so for white witnesses and AA suspects Stereotype threat and B responses to it may increase confirmation bias and tunnel vision Possible real-world solutions Interventions aimed at increasing police legitimacy Implement Community Policing Programs Teaching police officers to suppress stereotypes

Race section from Devine jury book

RACIAL ISSUES Similarity-leniency effect Similar defendants/jurors expected to be treated more leniently Black sheep effect Evidence of wrongdoing - jurors want to dissociate themselves from similar defendant Two types of studies done Archival analyses of actual death penalty cases Mock juror experiments Three meta-analyses done First (Sweeney & Haney, 1992) found a sig. effect of race on sentencing Second (Mazzella & Deingold, 1994) found no effects Third (Mitchell et al., 2005) found sig. effects of similarity leniency for white and blacks Racially diverse juries deliberated better than homogenous juries (Sommers, 2006) Field study by Taylor and Hosch (2004) found no support at the jury level for a similarity-leniency effect, outgroup-punitiveness effect, or a black sheet effect BUT field studies have confirmed influence of race in the sentencing phase of capital jury decision making Non-whites were less likely to prefer the death penalty than whites; # of white males on jury significantly associated with likelihood of jury opting for death penalty Summary Research consistent with similarity-leniency effect when it comes to race and the judgements of white and black jurors, particularly in capital cases where a black defendant is charged with killing a white victim

Batson v. Kentucky, 1986

RACIAL ISSUES Original case: Swain v. Alabama (1965) Black man indicted for second-degree burglary and receipt of stolen goods Judge used voir dire examination of the venire to conduct peremptory challenges Removed all four black persons; a petit jury of all whites was selected Ruled that didn't violate 6th or 14th amendments Did not violate Equal Protection Clause Will not exclude members from the venire due to race Burden on Defendant to prove that prosecution was discriminatory in elimination of members from the venire Purpose of current case Swain teachings/understanding Proof needed to prove that discrimination took place Prosecutor must "in case after case, whatever the circumstances, whatever the crime and whoever the defendant or the victim may be... with the result that no Negroes ever serve on petit juries" Swain understanding WRONG - Clarified version A single discriminatory act is not "immunized by the absence of such discrimination in the making of other comparable decisions" Defendant must Show that they are a member of a cognizable racial group and that the prosecutor has exercised peremptory challenges to remove members of that race from the venire Must rely solely on fact as to which there can be no dispute of discrimination Must show that these facts and any other relevant circumstances raise an inference that the prosecutor used that practice to exclude members of the venire based on their race Then the burden shifts to the State to come forward with a neutral explanation for challenging black jurors CANNOT be because they believed that the jurors would be more sympathetic towards the defendant due to a shared race (also part of the Equal Protection Clause) Goal of current decision: to enforce the mandate of equal protection, further the ends of justice, and to increase public respect for the CJ system and the law Issues with the current peremptory challenge system Defendants can't oppose unless challenges are obvious When defendants can oppose, trial courts face burden of assessing prosecutor's motives External vs internal biases hard to prove Proposed solution: ban both the prosecution's and the defendant's ability to exercise peremptory challenges Final finding Banned both the prosecution's and the defendant's ability to exercise peremptory challenges Due to equal protection principles

Findley, 2012

TUNNEL VISION Heuristics and logical fallacies that lead people to focus on a conclusion, select and filter information that points to that conclusion while ignoring evidence that points away from that conclusion Usually product of human condition; institutional or cultural pressures Very impactful in CJ system Feedback loop of evidence, created by tunnel vision Mistaken eyewitness ID -> coerced confession -> false informant testimony -> biased analyses -> bolsters confidence of eyewitness Tunnel vision can impact lawyers and judges too Cognitive distortions Confirmation Bias Belief perseverance Natural tendencies make people resistant to change even in the face of new evidence that wholly undermines their initial hypotheses Hindsight Bias (effect of outcome information on probability of an outcome) Suspect guilt Quality of evidence used to convict Reiteration effect Confidence of an assertion increases if the assertion is repeated Outcome Bias (effect of outcome information on evaluations of decision quality) Anchoring effects Role effects Conformity effects Experimenter effects External forces of Tunnel Vision Institutional pressures on police Performance standards Police culture and training Institutional pressures on prosecutors Selective funneling of information to prosecutors Lack of diagnostic feedback on accuracy of prosecutions Institutional pressures on defence Laws that limit inquiry, trials, and appeals Research Hard to simulate to study No direct studies Professionals in other fields experience Tunnel Vision though (found with bank personnel) Fingerprint experts Police officers in lie detection Interrogators during interviews People who are told to form a hypothesis of guilt early in evaluation of evidence show confirmation bias in investigative scenario Implications Appellate court Hindsight bias or Outcome bias Harmless error doctrine or ineffective counsel Possible Solutions Reforms Legal rules of procedure Police training and standards Incentives for police other than clearance and conviction rates Education of cognitive biases to all parties involved Research says not really effective BUT articulating reasons (for self and opposite party) is Consider and articulate counter arguments Solution: transparency Institutionalized counter-arguing Creating devils advocates Won't be fully effective against bias but It would help Expanded discovery in criminal cases So defense has a chance against biases Legal decisions more open and observable Both police and prosecutors

Hill et al., 2008

TUNNEL VISION Investigative interviews Reid is similar to coercive techniques used in other countries Police and Criminal Evidence Act 1984 (PACE)and National Package on Investigative Interview Training in England and Wales Even PACE not perfect - 10% of interviews in a study of 177 breached PACE Research Guilt assumption in police interviewing Confirmation bias Not really studied Snyder & Swann (1978) Hypothesis testing in social interaction People tend to ask hypothesis confirming questions, and interviewees tended to behave in ways that confirmed hyp Self-fulfilling prophecy effect Kassin, Goldstein, Savitsky (2003) Interrogating a mock crime with presumptions of innocence or guilt Given a list of questions to pick from Confound: In guilt presumptive condition, more questions and techniques were used than in the innocent condition Presumptions of guilt affected Type of questions and number of techniques used Suspect behavior during interview Perception of suspect by third party Current Experiment First study Whether an assumption of guilt affects questions asked, when free to generate questions Results Interviewers that presume guilt generate more guilt-presumptive questions, even when free to generate their own questions Confirms confirmation bias More confident in judgements of guilt/innocence when presumed suspect to be guilty Second study Whether questioning styles from study 1 influence confession or denial rates Results Guilt or innocence was associated with confession or denial Questioning-style was not enough to induce true or false confessions Increased feelings of pressure for confession Third study Effect of questioning styles on suspects' behavior (if a self-fulfilling prophecy occurs) Results Participants in guilt-presumptive questioning conditions were rated as more likely to be guilty due to their behaviors When innocent suspects responded to guilt-presumptive questions, the self-fulfilling prophecy effect was more pronounced Verbal response perceived as indicative of guilt Suspects perceived to be more nervous, defensive, and less plausible when responding to guilt-presumptive questions Raters were highly confident in erroneous judgements of guilt Shows that guilt-presumptive questioning alone is strong enough to produce a self-fulfilling prophecy effect Discussion Investigator Bias Effect Police have bias towards judgements of deceit Not correlated with accuracy High confidence, low accuracy Jonas et al. (2001) Significantly stronger preference for hypothesis confirming information when information is gathered sequentially Seeking more supporting information more likely than less conflicting information

Kersthol & Eikelboom, 2007

TUNNEL VISION Tunnel vision in group setting - groupthink Group reaches interpretation Individuals less likely to assess independently Crime analysts Look for patterns in data Information foraging Making sense phase Exist as outside second opinion, independent evaluators But if they know the investigation team's prior theory, does it still work? Decisions highly sensitive to previous information and perspective Even experts are susceptible to anchoring, confirmation bias Current EXP: Whether crime analysts are influenced by prior knowledge of the team's preferred interpretation and to what extent experience has a moderating effect Evaluate two different cases with two different tasks First task: interpret roles of various persons in a slave trade case Prior interpretation: one of women had a key role in organization Second task: generate alternative explanations for disappearance of a young woman case Prior interpretation: father played a major role Prior interpretations are plausible but not most likely scenario, to be able to tell if the interpretation affects decision making Compared less experience (7 mo) to more experienced (7 yr) Findings Prior knowledge compromises analysts independent analysis In both complex and simple cases Mentally "joined the team" Maybe due to temporarily treating a hypothesis as truth leading to confidence in the hypothesis? Those who read interpretation first mentioned same evidence as interpretation more than other evidence not included in interpretation Prior interpretation did not affect how analysts weighted information Evidence with most weight were those that contributed to narrative or "story" No moderating effect of expertise

O'Brien and Findeley, 2014

WRONGFUL CONVICTIONS OVERVIEW APPROACH - Chapter on psychological errors in criminal justice system MAIN FINDINGS - Most common source of error in criminal cases: - "Gross's Canonical List" - Eyewitness ID error - Flawed or false forensic science - False confessions - Perjury by informants - Prosecutorial misconduct - Inadequate defense counsel - Innate cognitive processes that can lead to decision errors: broadly called Tunnel Vision - Confirmation Bias - All throughout justice system - Belief Perseverance - Jurors being asked to disregard inadmissible evidence - Hindsight Bias - Likelihood of winning a case - Outcome Bias - Looking back at trial: that argument was inappropriate, but probably wouldn't have changed outcome - harmless error - Mindset Theory - More deliberate than confirmation bias Open-minded vs narrow-minded (not trying to solve the case anymore, just prove MY case) - Cognitive Dissonance - Prosecutor would never put away an innocent; but one of his convictions is exonerate by DNA. Justify by continuing to believe guilty, despite exoneration - GroupThink - Needs a relationship between parties, have negative consequences to diverging - Limitations on Human Memory - Estimator variables; Outside of CJ system's control - Stereotyping - Conviction - Racial Bias - Lineup construction, investigation - Other - Diffusion of Responsibility - Belief in the system/justice - Relationships between parties involved "Team Players" - Judges and prosecutors liking quick defense lawyers, prosecutor/police relations, etc - Conforming within the power hierarchy - Between colleagues, predecessors, superiors - Jurors feelings pressured to agree with others, assuming the charges are founded because they are being charged, etc (conviction as default instead of other way around) IMPORTANT TAKEAWAYS - Cognitive Phenomenon affect every part of the criminal justice system