The literature has theorized and highlighted the impact of cognitive and motivational biases on judgment making (Kruglanski and Azjen, 1983; Montibeller and von Winterfeldt, 2015). Cognitive biases are due to the limitations of the human being to process all the information, which lead to direct attention toward certain information and discard other that could be equally relevant (Nisbett and Ross, 1980). In sum, cognitive biases arise from limitations in human information-processing substituting an expected formal reasoning (for a review see Kruglanski and Azjen, 1983). Thus, in contexts in which judgment making must rest on formal reasoning, such as legal judgments, these biases should not have a place. However, research on sexual violence has shown that judgment making about it is influenced by myths about sexual assault (Lonsway and Fitzgerald, 1994), which serve as descriptive or prescriptive cognitive tools about the causes, contexts and consequences of sexual assaults, as well as perpetrators, victims and their interaction. These types of cognitive schemes allow access to heuristic representations of information to judgment making about sexual violence (McKimmie et al., 2020). Thus, individuals sharing these myths use them to deny, minimize, overgeneralize or justify the violence of men against women (Gerger et al., 2007), while favoring risky sexual behaviors (Álvarez-Muelas et al., 2020), and a different evaluation of the same behaviors by men and women (sexual double standard; Álvarez-Muelas et al., 2021), a contingency with a high prevalence in the Hispanic context (Martínez-Gómez et al., 2021). Conversely, the perception of the complainant as chaste, respectable or sober i.e., gender victim stereotypes (counter-myths) is related to the opposite trend (Schuller et al., 2010). For this purpose, judicial judgment makers, in line with the judicial reasoning (law of precedent), are impelled to use the assignment or not of credibility (reliability in scientific judgment making models; Kaplan et al., 1978) to the testimony of the complainant (Arce et al., 2000; Du Mont et al., 2003; Page, 2007, 2010; Schuller et al., 2010; Anders and Christopher, 2011; Hine and Murphy, 2017).

On the other hand, motivational biases are characterized by a tendency to form and hold beliefs that fulfill the needs of the individual or overestimate the perceived degree of controllability of the environment (Novo and Seijo, 2010). Among the motivational biases, attributional biases have been related (Burger, 1981), which are used as means for judgment making through the attribution of responsibility or credibility (judicial task). In judgment making about sexual assault, expectations persist about how a real victim of a sexual assault behaves, which are a breeding ground for attributional biases (Smith and Skinner, 2017) that are not present in other types of crimes not including the issue of consent that plays a critical role in rape cases (Bieneck and Krahé, 2011). Thus, these sources of bias are used to attribute responsibility to the complainant and aggressor, to attribute credibility to the testimony of the complainant (proof of charge, while the testimony of the accused is not evidence of the charge, so from its evaluation cannot be derived criminal responsibility) or to attribute to the alleged facts nature of a sexual assault (most sexual assault complaints have to be resolved on whether the facts are an assault or a consensual relationship).

Biases acquire a functional nature as the evidence is weak and lose it when the evidence is strong (Visher, 1987; Kahneman, 2011; Butterfield and Bitter, 2019; Nitschke et al., 2019). In crimes committed in the private sphere (e.g., domestic violence, sexual assault) there are, therefore, few media of burden of proof beyond the testimony of the complainant and the evaluation of the damage to the complainant (Arce, 2017). Hence, trials in these crimes are conducive to the manifestation of bias.

These types of cognitive and motivational biases affect not only judicial judgment making of laypersons (the vast majority of research has been carried out with jurors, i.e., laypersons; Schuller et al., 2010), but also of law professionals (Fitzmaurice and Pease, 1986; Saks and Kidd, 1986; Fariña et al., 2002; Arce et al., 2003). In the European and Anglo-Saxon context, the presence and impact of myths about sexual violence on police and judicial agent samples has been widely documented (Camplá et al., 2017; Smith and Skinner, 2017; Temkin et al., 2018). Nevertheless, not all have the same meaning in this type of population. Thus, one of the most prevalent myths reported in the population of professionals with procedural or judicial competences (Police, Prosecutors, Judges) is the one referring to false (deliberately fabricated)/unfounded (not deliberately fabricated, baseless, groundless) accusations, based on the belief that women allege for revenge, for profit or regret (Lisak et al., 2010; Lonsway, 2010; Ferguson and Malouff, 2016). As well, the Chilean Law of Precedent establishes that the testimony of the complainant is not sufficient evidence if there is some benefit, revenge or repentance in it. Hence, if this were the only proof against the accused, the judicial criterion would classify the case as evidenceless and it would be closed or archived, expanding in this group unfounded to evidenceless.

In any case, the estimation of the probability of false or unfounded accusations has been considered as a source of bias in trials in sexual assault crimes. In this regard, a meta-analytical review (Ferguson and Malouff, 2016) found a high inter-study variability and that the results were subject to the effect of moderators that could not be identified due to lack of studies (possibly moderators of the effects are the definitions of false and unfounded complaint and the type of population). While waiting to know these moderators, the lower (0.012) and upper (0.174) limits reported demarcate the estimates within normality (between the interval of the best estimates), with the lower and upper estimates being outside the normal range.

In Latin-American judicial setting, it has been argued that investigative actions and judicial judgment making in sexual assault cases may be contaminated by prejudices about sexual assault and the complainant (Salinas et al., 2015). Hence a quasi-experimental field study with Chilean judicial agents (i.e., law enforcement officers, correction officers, prosecutors, and judges), as professionals with procedural or judicial competences, to estimate the probability of false or unfounded complaints, the incidence of myths about sexual assaults, and the incidence and effects of motivational biases in judgment making was designed.

Regarding the incidence of myths about sexual assault, the results obtained with Chilean judicial agents showed, in general, a systematic tendency of disagreement with the myths about sexual assault. Bearing in mind the theoretical content categories of the scale (Gerger et al., 2007), they express disagreement with the denial of the scope of sexual violence, as well as with gender stereotypes about male sexuality, the beliefs that exonerate perpetrators of violence, and the naturalization of male coercion. At the same time, they are in favor of the demands of the victims and of the policies designed to address the effects of sexual violence. Nevertheless, there is no systematic trend of agreement or disagreement regarding the false accusation of sexual violence to obtain custody, and the interpretation of harmless gestures as sexual harassment. On the contrary, they participate in the myth of positive discrimination “receive more psychological support” rape victims compared to victims of armed robbery that are not based on the provision of greater support to victims of sexual assault. De facto, all Chilean victims of violent crimes (e.g., sexual assault, kidnapping, robbery with violence) receive legal assistance, psychological therapy and social support. In any case, the results do not endorse a generalized cognitive bias in this population contrary to those reporting sexual assault victimization, observed in other studies (Sleath and Bull, 2012; Smith and Skinner, 2012; McMillan, 2016; Hine and Murphy, 2017; Temkin et al., 2018).

However, the case study, necessary, as it deals with personal biases (De Neys and Bonnefon, 2013), indicates that a large part of the legal agents participate in the myths about sexual assaults, with agreement rates that oscillate between ± 25%, at +50%. In sum, although a systematic bias trend is not observed in the population studied, a significant and large prevalence of cases has been recorded (ranging from 1/4 to more than 50% of judicial agents, depending on the myths).

Myths are part of sources of informal reasoning (evidence is replaced by the myths), as opposed to formal reasoning (evidence based) that must support judicial judgments (Kruglanski and Azjen, 1983). In practice, myths are within the cognitive bias “preconceived ideas or theories” that predispose the individual to adopt uncertain ideas (e.g., myths) that guide judgment making via information processing strategies such as presumed covariation (e.g., correlation between myths and false or unfounded allegations), representativeness (e.g., overestimation of the probabilities of false or unfounded allegations related to myths) or causality (e.g., myths are the causes of false or unfounded allegations) (Fariña et al., 2002). These preconceptions maximize judgments based on myths, avoid discordant information, and lead to cognitive savings for the individual (Ross, 1977).

In relation to the estimation of false or unfounded reports (representativeness cognitive bias), it was found that the mean probability reported by the study population is significantly higher than the mean reported in the literature (0.052; Ferguson and Malouff, 2016), and is in the upper limit of the best estimates. In addition, the study of cases, according to the results of the meta-analytical review by Ferguson and Malouff (2016), has allowed in order to establish for legal agents a common overestimation of false or unfounded complaints, as well as a non-trivial underestimation, although the overestimation is significantly greater than the underestimation (Venema, 2016). In this sense, research has shown that high estimates are related to a lower allocation of credibility and receptivity toward those who report (Lonsway et al., 2009; Mennicke et al., 2014). Likewise, judicial agents with differentiated procedural functions (who activate the search for evidence or file the process), police and prosecutors, and operators without procedural functions (they do not make decisions about the process), gendarmes and judges, did not reveal differences in the estimation of false or unfounded complaints, which may condition the decision about a criminal prosecution or contribute to poor investigations (Mennicke et al., 2014; Hohl and Stanko, 2015; O’Neal et al., 2015; Carboné-López et al., 2016). In any case, it is remarkable that the assessment of the probability of false or unfounded complaints is conditioned in this population by the judicial criterion of subjective incredibility (Law of Precedent) that establishes that the testimony of the complainant is not sufficient proof of incrimination if he/she has any interest in the cause beyond the legitimate conviction of the accused (e.g., economic benefit, resentment, revenge, existence of a previous relationship).

Regarding attributional biases, the results confirmed that, for judgment making and, by extension, judicial decision-making, judicial agents use sources of informal reasoning (attributional biases) in contrast to the formal reasoning expected in this context. These biases facilitate paths of judgment both incriminating (i.e., attribution of responsibility to the accused, attribution of credibility to the complainant, attribution to the facts of a nature of rape) and exculpatory (i.e., attribution of responsibility to the complainant, attribution of the duty of prevention to the complainant). Furthermore, the magnitude of the effects of the biases in the reasoning is more than great. However, the biases in favor of incriminating the accused (attribution of responsibility to the accused, attribution of credibility to the complainant, attribution to the facts a nature of rape) have a greater weight than those that delegitimize the accusation (attribution of responsibility to the complainant, attribution of the duty of prevention to the complainant). Moreover, the results revealed that these have a smaller effect on known victims than on unknown ones. Succinctly, higher effects for attributional biases are for unknown perpetrators in both directions: to support incrimination (i.e., attribution of higher veracity to complainant testimony) and absolution/dismiss of the judicial proceeding (i.e., higher attribution of responsivity to complainant, higher attribution of capacity to prevent the incident to complainant). Nevertheless, the attribution of credibility to the complainant testimony has a higher incriminating evidence value than the attribution of responsivity and capacity to prevent the incident to the complainant for exonerating criminal responsibility (Arce et al., 2003). In this way, the existence of a close link between the complainant and the accused requires a greater burden of proof, since less verisimilitude is attributed to the complainant testimony (McKimmie et al., 2014), more responsibility is attributed to the complainant in the facts and more ability to have prevented the incident (Hohl and Stanko, 2015; Hine and Murphy, 2017). In the absence of such an increased burden of proof, these attributional biases would predispose judgment making toward the absolution of the prosecuted or the dismiss of the judicial proceeding. Paradoxically, this is related to a lower probability of reporting and abandoning the relationship (Garrido-Macías et al., 2020) and, by extension, less judicial protection for victims of known aggressors. However, the case study warns that around 1/3 of the judicial agents (34.0, 30.2, and 28.8%, respectively), would attribute, respectively, more responsibility, truthfulness, and prevention capacity to the complainant when the perpetrator is known (statistical model error).

Motivational attribution biases refer to a tendency to form and hold beliefs that conform to the needs of the individual, in this case, judgment making and the subsequent decision-making, and manifest when the legal evidence is insufficient or weak (week cases; Butterfield and Bitter, 2019), being irrelevant in strong evidence cases (Visher, 1987; Kahneman, 2011). Under this contingency, judicial agents, in judicial judgment and decision making, must be guided by strict compliance with the principle of presumption of innocence (Article 11.1 of the Universal Declaration of Human Rights; United Nations, 1948), which implies that none innocent person may be classified as guilty, so the attributional biases supporting the absolution of the accused of the crime would support the motivation of the procedural action taken or the judicial resolution. However, although the judicial resolution or procedural action (i.e., prosecutorial decision making) executed would be correct, it would be based on informal reasoning, which would deviate from normative judicial reasoning (i.e., evidenceless, unfounded). On the contrary, in cases of weak or insufficient legal evidence (burden of proof), resorting to attributional biases to support the case would not only be inadmissible in terms of reasoning (motivation of the actions or judicial resolutions) as it is informal in the face of the expected formal, but also contrary to law (judicial error). In any case, attributional biases, such as irrational beliefs, provide such a high level of support (effect size greater than large) that they give the subject a guarantee of certainty and efficiency in judgment making (Perry, 1988).

In conclusion, although generally and in population terms the effect of biases in judgment making contrary to complainant is not observed, in the case study it was found that a large number of judicial agents participated of biased routes against the complainant. Moreover, this type of bias in judgment making does not occur in other crimes (Bieneck and Krahé, 2011). This research found that in rape cases was attributed more blame to the victim and less blame to the perpetrator compared with robbery cases. Thus, as these sources of bias in judgment making are unconscious for judgment makers and ways of informal reasoning, judicial agents should be trained to control the effects of these sources of bias (Bartels, 2010), promoting debiasing, i.e., substituting informal reasoning (judgment making sustained on biases against the complainant) by formal reasoning sources (evidence, procedural rules, charge of the proof) (Butterfield and Bitter, 2019). In sum, the training and specialization of judicial agents (e.g., courts specialized in sexual assaults, training police forces to obtain the statement from complainants of sexual assault) in sexual violence against women cases is necessary (Barn and Powers, 2021; Gancedo et al., 2021b); so that, from an orientation of Therapeutic Justice, they can mediate the wellbeing of the victims (Cattaneo and Goodman, 2010; Camplá et al., 2020; Novo et al., 2020).

The results of this study are not generalizable to other types of populations, since judicial agents are determined in their judgments by procedural and legal rules and Chilean Law of Precedent. Likewise, caution must be taken in generalizing judicial agents from judicial contexts other than Chile, since the case law may not be equivalent. As participation was voluntary, the results do not represent strictly the population. Finally, the manifestation of these biases can only be generalized to cases of insufficient evidence.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethical review and approval was not required for the study on human participants in accordance with the local legislation and institutional requirements. The patients/participants provided their written informed consent to participate in this study.

All authors listed have made a substantial, direct, and intellectual contribution to the work, and approved it for publication.