Why eyewitnesses fail

A call for open science in forensics

The modern canon of open science consists of five "schools of thought" that justify unfettered access to the fruits of scientific research: i) public engagement, ii) democratic right of access, iii) efficiency of knowledge gain, iv) shared technology, and v) better assessment of impact. Here, we introduce a sixth school: due process. Due process under the law includes a right to "discovery" by a defendant of potentially exculpatory evidence held by the prosecution. When such evidence is scientific, due process becomes a Constitutional mandate for open science. To illustrate the significance of this new school, we present a case study from forensics, which centers on a federally funded investigation that reports summary statistics indicating that identification decisions made by forensic firearms examiners are highly accurate. Because of growing concern about validity of forensic methods, the larger scientific community called for public release of the complete analyzable dataset for independent audit and verification. Those in possession of the data opposed release for three years while summary statistics were used by prosecutors to gain admissibility of evidence in criminal trials. Those statistics paint an incomplete picture and hint at flaws in experimental design and analysis. Under the circumstances, withholding the underlying data in a criminal proceeding violates due process. Following the successful open-science model of drug validity testing through "clinical trials," which place strict requirements on experimental design and timing of data release, we argue for registered and open "forensic trials" to ensure transparency and accountability.

Memory accuracy, suggestibility and credibility in investigative interviews with native and non-native eyewitnesses

Legal practitioners sometimes obtain eyewitness testimonies from non-native language speakers, yet few studies examine the effects of language in investigative interviews. Here, we investigate how testifying in a non-native vs. native language affects memory accuracy, susceptibility to suggestions, and witnesses perceived credibility. After viewing a mock-crime film, participants in Study 1 (N = 121) testified through (1) free recall, (2) cued recall and (3) the Gudjonsson Suggestibility Scales adapted to the crime scenario used in the present study either in their native (Swedish) or a non-native (English) language. They also rated their confidence in their memories, self-perceived credibility and cognitive effort. Native and non-native witnesses did not differ in memory accuracy, susceptibility to suggestions, self-rated credibility or cognitive effort. Non-native (vs. native) speakers did however report lower confidence in their memories. In Study 2, another group of participants (N = 202) were presented with the testimonies from Study 1, and judged witnesses’ credibility. Non-native witnesses were judged as less credible than native speakers. Thus, while the lower confidence exhibited by non-native eyewitnesses did not correspond to their actual memory accuracy, it influenced observers’ judgments of their performance. The results provide important knowledge for legal practices when evaluating the reliability of testimonies from non-native vs. native speaking eyewitnesses.

Recent advances in Forensic DNA Phenotyping of appearance, ancestry and age

Forensic DNA Phenotyping (FDP) comprises the prediction of a person's externally visible characteristics regarding appearance, biogeographic ancestry and age from DNA of crime scene samples, to provide investigative leads to help find unknown perpetrators that cannot be identified with forensic STR-profiling. In recent years, FDP has advanced considerably in all of its three components, which we summarize in this review article. Appearance prediction from DNA has broadened beyond eye, hair and skin color to additionally comprise other traits such as eyebrow color, freckles, hair structure, hair loss in men, and tall stature. Biogeographic ancestry inference from DNA has progressed from continental ancestry to sub-continental ancestry detection and the resolving of co-ancestry patterns in genetically admixed individuals. Age estimation from DNA has widened beyond blood to more somatic tissues such as saliva and bones as well as new markers and tools for semen. Technological progress has allowed forensically suitable DNA technology with largely increased multiplex capacity for the simultaneous analysis of hundreds of DNA predictors with targeted massively parallel sequencing (MPS). Forensically validated MPS-based FDP tools for predicting from crime scene DNA i) several appearance traits, ii) multi-regional ancestry, iii) several appearance traits together with multi-regional ancestry, and iv) age from different tissue types, are already available. Despite recent advances that will likely increase the impact of FDP in criminal casework in the near future, moving reliable appearance, ancestry and age prediction from crime scene DNA to the level of detail and accuracy police investigators may desire, requires further intensified scientific research together with technical developments and forensic validations as well as the necessary funding.

Acute alcohol intoxication and alcohol expectancy effects on women's memory for consensual and non-consensual sexual activity

Objective

To test whether acute alcohol intoxication and alcohol expectancy affects how accurately women remember consensual and non-consensual sexual activity that occurred during an interactive hypothetical dating scenario.

Design

A balanced placebo randomized study that varied alcohol dose (mean Breath Alcohol Content; BrAC = 0.06%) and alcohol expectancy prior to participants encoding a hypothetical interactive rape scenario was implemented. Participants could elect to consent to sexual activity with a male partner in the hypothetical scenario. If they stopped consenting, non-consensual sexual intercourse (i.e., rape) was described. Seven days later, participants' memory for consensual and non-consensual sexual activity in the scenario was tested.

Main outcome measures

Memory accuracy, confidence, and feelings of intoxication.

Results

A total of 90 females (M age = 20.5, SD = 2.2) were tested regarding their memory accuracy for the consensual and non-consensual sexual activities in the scenario. A multi-level logistic regression predicting memory accuracy for the perpetrator's behaviors during the rape indicated no effect of alcohol intoxication. However, a main effect of alcohol expectancy was found, whereby participants who expected to consume alcohol, compared to those who did not, recalled the perpetrator's behaviors during the rape more accurately. A second regression predicting memory accuracy for consensual sexual activity found no main effects for alcohol intoxication or alcohol expectancy. Participants recalled consensual sexual activity with a high degree of accuracy. Calibration analyses indicated that accuracy increased with confidence level, regardless of intoxication level or alcohol expectancy condition, but that women tended to be overconfident in general.

Conclusion

This study provides an important test of how accurately women remember consensual and non-consensual sexual activities. The accuracy of this information is important for forensic medical examinations and police investigations following an allegation of sexual assault. Increased memory accuracy was found for offence details when participants expected to consume alcohol, suggesting there may be important differences in attentional processes (e.g., hypervigilance) depending on whether threat is present. Further research is necessary to investigate memory for sexual violence in real-world settings and to test methods for ascertaining the most complete and reliable accounts.

Scientific Thinking About Legal Truth

In the criminal process, the fact finders assess the validity of impressions reported by witnesses based on their perceptions and determine what has happened in reality. However, these impressions are not subject to any external validity check. The Innocence Project revealed the failure of this subjective method and showed how it can lead to innocent convictions. The legal literature has examined ways to manage the risk of mistakes, but these ways are inconsistent with the scientific understanding of the need for external validity measurements, suggesting the need for new ways of thinking about the legal search for truth and justice.

Identifying criminals: No biasing effect of criminal context on recalled threat

To date, it is still unclear whether there is a systematic pattern in the errors made in eyewitness recall and whether certain features of a person are more likely to lead to false identification. Moreover, we also do not know the extent of systematic errors impacting identification of a person from their body rather than solely their face. To address this, based on the contextual model of eyewitness identification (CMEI; Osborne & Davies, 2014, Applied Cognitive Psychology, 28[3], 392-402), we hypothesized that having framed a target as a perpetrator of a violent crime, participants would recall that target person as appearing more like a stereotypical criminal (i.e., more threatening). In three separate experiments, participants were first presented with either no frame, a neutral frame, or a criminal frame (perpetrators of a violent crime) accompanying a target (either a face or body). Participants were then asked to identify the original target from a selection of people that varied in facial threat or body musculature. Contrary to our hypotheses, we found no evidence of bias. However, identification accuracy was highest for the most threatening target bodies high in musculature, as well as bodies paired with detailed neutral contextual information. Overall, these findings suggest that while no systematic bias exists in the recall of criminal bodies, the nature of the body itself and the context in which it is presented can significantly impact identification accuracy.

Fooling ourselves and others: confirmation bias and the trustworthiness of qualitative research – Part 2 (cross-examining the dismissals)

Purpose

The purpose of this paper is to describe and critique ways in which the threats from confirmation bias have been rejected.

Design/methodology/approach

Dismissals of the existence of, or threats from, confirmation bias are identified from a review of literature across a very wide range of disciplines. The dismissals are robustly examined.

Findings

The dismissals are categorised as: (1) radical scepticism (2) consequentialism: and (3) denial. Each type of dismissal, it is argued, is flawed.

Originality/value

The three-fold structuring of confirmation bias dismissal is novel. In addition to drawing from organisation, management and wider social science literature, the article also uses arguments and examples from the creative arts.

He did it! Or did I just see him on Twitter? Social media influence on eyewitness identification

Eyewitness identification is fallible, and suggestive post-event information is known to facilitate error; however, whether social media valence affects identification decisions is unknown. After viewing crime videos of various race perpetrators, participants saw post-event Twitter photos of the perpetrator or a foil that varied in valence. Participants attempted identification from a lineup including both individuals and rated the confidence and source (i.e., video, Twitter) of their selection. Results showed that Twitter photos of the perpetrator increased the likelihood of accurate identification and related confidence, whereas seeing the foil reduced the likelihood of a correct identification and related confidence. Remembering the perpetrator from the crime influenced correct identification, while remembering other incorrect sources (e.g., Twitter) only lead to misidentification. Twitter valence and perpetrator race did not impact outcomes. Results suggest that difficulty in identifying a perpetrator is underpinned by source monitoring confusion which is exacerbated by viewing social media that includes innocent suspects.

Social and environmental factors influencing contemporary cases of wolf aggression towards people in Poland

Understanding factors and mechanisms causing large carnivore aggression towards people is crucial for their conservation in modern human-dominated landscapes. We present detailed descriptions of wolf attacks on people in early summer 2018 in two areas of Poland, analysis of behaviour, fitness and origin of individuals responsible for the attacks and management actions undertaken by local communities as well as governmental bodies and non-governmental organisations. We show that attacks were caused by yearling wolves (13-month-old male and 14-month-old female) originating from local packs, which appeared near households several months prior to incidents. Both individuals were positively food-conditioned and showed increasing habituation caused by irresponsible behaviour of people such as long-lasting intentional feeding or illegal keeping. Post mortem necropsy revealed obesity and serious abnormality of spleen (polysplenia) in the problem female wolf that negatively impacted her fitness and was the most likely reason for her presence near households and feeding on leftovers. Despite prolonged observations of these wolves very close to human settlements (less than 30 m), no mitigation actions were undertaken until the attacks, after which both individuals were killed. We provide several recommendations to avoid aggressive human-carnivore encounters.

Towards Forensic DNA Phenotyping for Predicting Visible Traits in Dogs

The popularity of dogs as human companions explains why these pets regularly come into focus in forensic cases such as bite attacks or accidents. Canine evidence, e.g., dog hairs, can also act as a link between the victim and suspect in a crime case due to the close contact between dogs and their owners. In line with human DNA identification, dog individualization from crime scene evidence is mainly based on the analysis of short tandem repeat (STR) markers. However, when the DNA profile does not match a reference, additional information regarding the appearance of the dog may provide substantial intelligence value. Key features of the dog's appearance, such as the body size and coat colour are well-recognizable and easy to describe even to non-dog experts, including most investigating officers and eyewitnesses. Therefore, it is reasonable to complement eyewitnesses' testimonies with externally visible traits predicted from associated canine DNA samples. Here, the feasibility and suitability of canine DNA phenotyping is explored from scratch in the form of a proof of concept study. To predict the overall appearance of an unknown dog from its DNA as accurately as possible, the following six traits were chosen: (1) coat colour, (2) coat pattern, (3) coat structure, (4) body size, (5) ear shape, and (6) tail length. A total of 21 genetic markers known for high predicting values for these traits were selected from previously published datasets, comprising 15 SNPs and six INDELS. Three of them belonged to SINE insertions. The experiments were designed in three phases. In the first two stages, the performance of the markers was tested on DNA samples from dogs with well-documented physical characteristics from different breeds. The final blind test, including dogs with initially withheld appearance information, showed that the majority of the selected markers allowed to develop composite sketches, providing a realistic impression of the tested dogs. We regard this study as the first attempt to evaluate the possibilities and limitations of forensic canine DNA phenotyping.

A perceptual scaling approach to eyewitness identification

Eyewitness misidentification accounts for 70% of verified erroneous convictions. To address this alarming phenomenon, research has focused on factors that influence likelihood of correct identification, such as the manner in which a lineup is conducted. Traditional lineups rely on overt eyewitness responses that confound two covert factors: strength of recognition memory and the criterion for deciding what memory strength is sufficient for identification. Here we describe a lineup that permits estimation of memory strength independent of decision criterion. Our procedure employs powerful techniques developed in studies of perception and memory: perceptual scaling and signal detection analysis. Using these tools, we scale memory strengths elicited by lineup faces, and quantify performance of a binary classifier tasked with distinguishing perpetrator from innocent suspect. This approach reveals structure of memory inaccessible using traditional lineups and renders accurate identifications uninfluenced by decision bias. The approach furthermore yields a quantitative index of individual eyewitness performance.

Blinding or information control in diagnosis: could it reduce errors in clinical decision-making?

Background Clinical medicine has long recognized the potential for cognitive bias in the development of new treatments, and in response developed a tradition of blinding both clinicians and patients to address this specific concern. Although cognitive biases have been shown to exist which impact the accuracy of clinical diagnosis, blinding the diagnostician to potentially misleading information has received little attention as a possible solution. Recently, within the forensic sciences, the control of contextual information (i.e. information apart from the objective test results) has been studied as a technique to reduce errors. We consider the applicability of this technique to clinical medicine. Content This article briefly describes the empirical research examining cognitive biases arising from context which impact clinical diagnosis. We then review the recent awakening of forensic sciences to the serious effects of misleading information. Comparing the approaches, we discuss whether blinding to contextual information might (and in what circumstances) reduce clinical errors. Summary and outlook Substantial research indicates contextual information plays a significant role in diagnostic error and conclusions across several medical specialties. The forensic sciences may provide a useful model for the control of potentially misleading information in diagnosis. A conceptual analog of the forensic blinding process (the "agnostic" first reading) may be applicable to diagnostic investigations such as imaging, microscopic tissue examinations and waveform recognition. An "agnostic" approach, where the first reading occurs with minimal clinical referral information, but is followed by incorporation of the clinical history and reinterpretation, has the potential to reduce errors.

Modelling the effects of crime type and evidence on judgments about guilt

Concerns over wrongful convictions have spurred an increased focus on understanding criminal justice decision-making. This study describes an experimental approach that complements conventional mock-juror experiments and case studies by providing a rapid, high-throughput screen for identifying preconceptions and biases that can influence how jurors and lawyers evaluate evidence in criminal cases. The approach combines an experimental decision task derived from marketing research with statistical modeling to explore how subjects evaluate the strength of the case against a defendant. The results show that, in the absence of explicit information about potential error rates or objective reliability, subjects tend to overweight widely used types of forensic evidence, but give much less weight than expected to a defendant's criminal history. Notably, for mock jurors, the type of crime also biases their confidence in guilt independent of the evidence. This bias is positively correlated with the seriousness of the crime. For practicing prosecutors and other lawyers, the crime-type bias is much smaller, yet still correlates with the seriousness of the crime.

Diagnosing Crime and Diagnosing Disease-II: Visual Pattern Perception and Diagnostic Accuracy

Previously, we reviewed how general cognitive processes might be susceptible to bias across both forensic and clinical fields, and how interdisciplinary comparisons could reduce error. We discuss several examples of clinical tasks which are heavily dependent on visual processing, comparing them to eyewitness identification (EI). We review the "constructive" nature of visual processing, and how contextual factors influence both medical experts and witnesses in decision making and recall. Overall, studies suggest common cognitive factors uniting these visual tasks, in both their strengths and shortcomings. Recently forensic sciences have advocated reducing errors by identifying and controlling nonrelevant information. Such efforts could effectively assist medical diagnosis. We suggest potential remedies for cognitive bias in these tasks. These can generalize across the clinical and forensic domains, including controlling the sequencing of contextual factors. One solution is an agnostic primary reading before incorporation of a complete history and interpretation.