Logo: to the web site of Uppsala University

Predictability of legal decisions is usually considered a prerequisite for the rule of law, following the maxim ‘like cases should be treated alike’. Yet, this presupposes that the case outcome can be predicted based on the merits of the case, rather than other factors. The purpose of this study was to test whether and to what extent legal decisions on petitions for new criminal trials can be predicted on the basis of other fairly superficial criteria that one could access without even reading the case file, e.g. which Court decided, whether the applicant had legal representation etc. To this end, all petitions for new criminal trials submitted to the Swedish Supreme Court and the six Courts of Appeal in the time period 2010–2020 (n = 3915) were reviewed. This data formed the basis of a regression model which was then used to predict decisions regarding petitions in 2021. On the basis of access to legal representation and crime type, the regression model predicted accurately 100 % of the decisions made in 2021. This raises questions about the evidentiary basis for the decisions and also the role of judges in situations where their decisions are fully predictable.

When crime scene investigators (CSIs) encounter crime scenes with large volumes of blood, some selection and prioritization is often needed, and this will impact on what blood is and is not available for forensic analysis. What factors influence CSIs decision making process is largely unknown. This study examines the effects of awareness of limited resources and irrelevant contextual case information indicating either a homicide or a suicide on CSIs collection of blood traces. To this end, two scenario-based experiments with CSIs and novices were conducted. Overall, the results suggest that even when CSIs decisions are made under identical conditions, their trace selection varies both when it comes to numbers and locations. Furthermore, awareness of limited resources made CSIs collect fewer traces and their selections also varied following the contextual case infor-mation, showing similarities and differences with novices. Since blood traces can be used to establish both ac-tivity and identity the findings can have important implications for the subsequent investigation as well as trial.

Confirmation bias is a tendency to search selectively for and emphasize information that is consistent with a hypothesis or preferred conclusion. At the same time, opposing information is ignored or interpreted in ways that do not threaten the predetermined conclusion. Importantly, confirmation bias is a more or less subconscious phenomenon. Hence, legal actors and criminal practitioners who are expected to be objective, who may make genuine efforts to be objective, and who also perceive of themselves as objective, may still display a confirmation bias. This includes police officers, crime scene investigators, forensic analysts, pathologists, prosecutors, and judges, and the range of distinct decision-making tasks that these actors undertake; for example, interviews with suspects, witnesses, plaintiffs, crime scene investigations, autopsies, decisions about whether to press charges, and whether to convict. Since confirmation bias can be present at all stages of criminal investigations and proceedings, it constitutes a serious risk of error, for example resulting in wrongful convictions and wrongful acquittals. It is therefore essential to find and use bias mitigation measures, and a good starting point is the explanations of this bias provided in cognitive, emotional and motivational, social and organizational psychology. This includes measures such as changing decision-maker, contextual information management, linear sequential unmasking, and structured evaluations of the evidence.

Medical opinions are often essential evidence in criminal cases but relatively little is known about the factors that impact forensic doctors' decision making. This research examines the role and impact of having an alternative hypothesis while forming a medical opinion. A scenario-based experiment with forensic doctors (n = 20) was conducted. In two out of three scenarios, the existence of alternative hypotheses impacted the actual opinions reached, the confidence in the judgments and the perceived consistency with the plaintiff hypothesis. Investigative and legal actors should be aware of the possibility of biases and importance of having alternative hypotheses when requesting and evaluating medical opinions.

The most common purpose of setting up a suspect lineup is to test the durability of a guilt hypothesis. However, the validity of this test is directly dependent on how the lineup is set up, conducted and evaluated. Even seemingly subtle factors such as the number of times lineup photos are displayed to witnesses can impact on the risk of error, both false positives and false negatives. In an applied setting, a decision by investigators to “try again” and go for another lap of displaying the photos, for example when the witness did not identify anyone in the first lap, can be due to a genuine will to prevent false negatives. For example, investigators may believe that the witness recognizes the suspect as the perpetrator but needs another lap to be confident enough to make an identification. However, the perception that there is a risk of a false negative can stem from a confirmation bias, that is, that investigators maintain their belief that they have identified the right suspect even in the face of information that potentially challenges that hypothesis (that the witness does not recognize the suspect as the perpetrator).

The purpose of this research was to test whether and how witness accuracy and confidence are impacted by the number of laps used in a sequential photo lineup, comparing specifically one lap (original version) to two laps (two laps version). To this end, a scenario-based experiment was conducted in which participants (n = 164) were shown a film clip of a suspected rape and 1 week later suspect lineups (both target absent and present lineup) were conducted with the witnesses.

Overall, the results suggest that witnesses tested with the original version more rarely erroneously identified someone when the perpetrator was absent, compared to when the two laps version was used. Yet, the ability to identify the perpetrator (when the perpetrator was present) was not significantly different among participants tested with the different methods. Overall, witnesses tested with the two laps version stated a higher degree of confidence in their decisions than did witnesses tested with the original version. The average deviation between stated confidence level and accuracy was somewhat, although not significantly, higher with the two laps version than with the original version.

Hence, while investigators may fear that there is a substantial risk of a false negative if they do not continue with another lap after a witness has been unable to identify the suspect after the first lap, the present results in fact suggest that a second lap does not significantly reduce the risk of a false negative. However, a second lap does significantly increase the risk of a false positive. Taken together, the results imply that the original version of a sequential lineup (one lap) is a better option than the two laps version, although none of the versions are fool-proof.

Between Expert Reliability refers to the extent to which different experts examining identical evidence make the same observations and reach the same conclusions. Some areas of expert decision making have been shown to entail questions with relatively low Between Expert Reliability, but the disagreement between experts is not always communicated to the legal actors forming decisions on the basis of the expert evidence. In this paper, we discuss the issues of Between Expert Reliability in legal proceedings, using forensic age estimations as a case study. Across national as well international jurisdictions, there is large variation in which experts are hired to conduct age estimations as well as the methods they use. Simultaneously, age estimations can be fully decisive for outcomes e.g. in asylum law and criminal law. Using datasets obtained from the Swedish legal context, we identify that radiologists and odontologists examining knees or teeth images to estimate age seem to disagree within their own disciplines (radiologist 1 v. radiologist 2 or odontologist 1 v. odontologist 2) as well as across different disciplines (radiologist v. odontologist) relatively often. This may have large implications e.g. in cases where only one expert from the respective field is involved. The paper discusses appropriate ways for legal actors to deal with the possibility of lacking Between Expert Reliability. This is indeed a challenging task provided that legal actors are legal experts but not necessarily scientific experts.

Man får utgå från att domaren är en självständigt tänkande individ. Frågan är då om denna individ tänker likadant som några tusentals andra individer som utövar domaryrket. Eller är det kanske så att rättstillämpning endast är en rent mekanisk verksamhet där fakta stoppas in i ena ändan och ett och samma resultat kommer ut i det andra oavsett vem som sköter processen. Frågor rörande detta behandlar Moa Lidén i denna artikel.