Culture, trauma, and memory in investigative interviews

ABSTRACT

Police investigators, immigration officials, and other investigative interviewers often interview individuals from different cultural backgrounds about potentially traumatising events. Much of the work on the impact of negative life events on memory has overlooked cultural differences. In this article, we integrate insights from legal, clinical, and cross-cultural psychology to shed light on cross-cultural investigative interviews about negative life events. We review how negative life events may be experienced and expressed differently around the world, highlighting the limitations of the Western ‘trauma’ model. Next, we consider how culture and negative life events may interact to influence memory reporting in investigative interviews. We identify barriers to disclosure and effective communication in interviews, including the role of interpreters. Finally, we propose how research findings on culture, trauma, and memory can be incorporated into the recently adopted Méndez Principles for investigative interviewing.

KEYWORDS:

• Cultural differences
• cross-cultural communication
• traumatic memories
• investigative interviewing
• eyewitness memory

The most important requirement in any investigative interview, whether it be with an eyewitness or asylum seeker, is for the interviewee to provide a detailed, coherent, and comprehensive account to the interviewer in their own words. Investigative interviewers often need to gather information from interviewees who have experienced potentially traumatic events. This includes a police officer interviewing a rape survivor, a judge interviewing a witness of genocide, and an immigration official interviewing an asylum seeker who fears persecution in their home country. A complicating factor in such interviews is that the interviewer and the interviewee are often from different cultural backgrounds. Because trauma can be experienced and expressed differently across cultures, this can influence how individuals come across in interviews, what they remember, and what they disclose—a recipe for misunderstandings and miscommunication. We review the literature from legal, clinical, and cross-cultural psychology to shed light on how culture and potential trauma may interact to influence memory, disclosure, and communication in cross-cultural investigative interviews.

In the present article, we first reflect upon key terms that are ubiquitous in modern discourse, yet not necessarily uniformly understood or clearly defined (Definitions of culture and trauma). After defining how we use these terms, we explain cultural differences in how negative life events are experienced (The experience of trauma) and in how reactions to such events are expressed (The expression of trauma). We then move on to a crucial component of investigative interviewing, namely, how interviewees from different cultures remember negative life events (Memory for negative life events). Yet, what an interviewee reports in an investigative interview is not only determined by what they remember, but also by what they choose to share (Barriers to disclosure) and how the interviewer communicates with them (Cross-cultural communication). This becomes even more complicated when an interpreter is involved (Interpreters in investigative interviews). In the final section (Guidance for cross-cultural interviewers), we present recommendations that incorporate the reviewed research findings into the six Principles on Effective Interviewing for Investigations and Information Gathering (Méndez et al., 2021), also known as the Méndez Principles, recently adopted by the United Nations. These include guidance on the foundations, practice, vulnerability, training, accountability, and implementation of interviewing, with the aim of moving away from coercive interrogations and toward rapport-based interviews.

Definitions of culture and trauma

The term ‘culture’ has many different definitions, yet is often left undefined. For example, of the 87 articles included in Drenk et al.’s (2023) literature review on culture in witness testimony, only six defined culture. We follow Marsella and Yamada’s (2010) definition of culture as ‘shared learned behavior and meanings that are socially transmitted for purposes of adjustment and adaptation … represented externally in artifacts (e.g. food, clothing, music), roles (e.g. the social formation), and institutions (e.g. family, government)’ (p. 105). As observed by Marsella and Yamada, ‘even mental disorders must vary across cultures because they cannot be separated from cultural experience’ (p. 105). The dominant theoretical framework for exploring cultural differences in emotions, cognitions, and motivations is how one positions or perceives oneself in society, known as self-construal (Markus & Kitayama, 1991). Markus and Kitayama argue that in individualistic societies, the self is typically construed in terms of one’s own unique attributes, qualities, and dispositions (independent self-construal). In contrast, in collectivistic societies, the self is seen more in relation to and dependent on one’s social grouping or community (interdependent self-construal). When interpreting differences in behaviour, the literature has focused almost exclusively on comparing North American and Western European individualistic cultures to East Asian collectivistic cultures (Anakwah et al., 2020; Veillard, 2017). We argue that researchers need to develop a more nuanced understanding of differences in beliefs and practices within countries and regions to avoid making generalisations based on individualism-collectivism alone (see also Anakwah et al., 2020; Uskul & Cross, 2019). We acknowledge at the outset that many differences exist within cultures as well as between (see Fischer & Poortinga, 2018) and that these may impact how negative events are experienced and disclosed in investigative interviews.

The use of culture-bound concepts such as ‘distress’ and ‘trauma’ to characterise the cognitive and emotional state of interviewees is also potentially problematic within the literature on investigative interviewing. ‘Trauma’ can be used in the widest sense to describe negative life events as well as reactions to such events, or can be more narrowly defined in relation to post-traumatic stress disorder (PTSD). As a psychiatric diagnosis, PTSD was first included in the third edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III; American Psychiatric Association, 1980) and in the tenth revision of the International Statistical Classification of Diseases and Related Health Problems (ICD-10; World Health Organisation, 1990) and is still used in these classifications. Both specify exposure to an extremely threatening or horrific event, as well as the requirement to re-experience this event, avoid reminders of it, and remain hypervigilant to current threat. Additionally, there is often overlap between diagnoses. For example, while PTSD is one of the most common diagnoses reported in asylum-seeking populations, depression and anxiety disorders are also common, along with physical complaints of headaches and pain (Song & Teichholtz, 2019; Westergaard et al., 2023).

In modern Western mental health discourse, many serious negative life events are increasingly labelled as ‘traumatic’, which can have detrimental consequences. First, such labelling may make survivors feel like they should be experiencing mental health problems, which could become a self-fulfilling prophecy (Summerfield, 2004). Second, medicalising and labelling human reactions as a disorder locates the problem in the individual and obscures the focus on social, ethical, and moral implications (Hinton & Lewis-Fernández, 2011). As Gilbert (2008) notes, ‘just because emotional reactions to distressing circumstances can be found worldwide does not necessarily mean that they mean the same thing for people everywhere’ (p. 7). We therefore follow Bourke’s (2022) recommendation to avoid the term ‘traumatic event’, since it makes assumptions about an individual’s response to adverse events, and instead use the term ‘negative life event’. In other words, we challenge the assumption that a negative life event routinely results in psychological trauma.

The experience of trauma

Experiences likely to result in traumatisation in one culture may not have lasting mental health effects in other cultures (Bourke, 2022; Hunt, 2010; Summerfield, 2004). For example, in one study, Japanese war veterans appeared to be protected against PTSD symptoms due to cultural values of accepting one’s fate and enduring suffering (Friedman et al., 2004). Similarly, in societies where sexual violence is rampant (e.g. during the Rwandan genocide and the Guatemalan civil conflict), the response to rape may be one of resilience rather than one of ‘trauma’ (Bourke, 2022). Even in different time periods in Western cultures, negative life events were experienced differently. For example, a 1966 disaster in a Welsh village that left 144 schoolchildren and teachers dead, did not seem to result in lasting mental health problems for the survivors, despite the tragic nature of the event (Summerfield, 2004). Summerfield speculated that medicalized and psychologised ways of thinking were unfamiliar in Welsh popular culture at the time. Today, people in Western cultures would perhaps assume such an event would deeply traumatise survivors for life, and that assumption could work against their mental healing.

In a similar vein, the personal significance and interpretation (including the labelling) of an event is determined by individual and cultural factors. Cross-cultural differences in shame are apparent in the findings of Heim et al. (2022), who found that embarrassing events elicited combined guilt and shame responses in an American sample, but shame alone in a Malay sample. Moreover, even though the sense of shame associated with becoming a victim of sexual abuse is prevalent in many cultures, appraisals and social meanings change over time (Bourke, 2022). For example, the recent widespread ‘#MeToo’ movement has significantly impacted current popular discourse around sexual abuse, by increasing transparency about people’s experiences and providing a way of expressing empathy. This movement was inspired by the story of a young girl ‘confessing’ that she had been ‘guilty of the crime of having been forced’ to have sex with her mother’s partner (Bourke, 2022, p. 23). Thanks to the movement, survivors today may feel less isolated and ashamed, and more likely to blame the abuser rather than themselves.

Shame can also be used in culturally specific ways by perpetrators to increase the extent of traumatisation (Hinton & Good, 2016). For example, in Rwandan culture, a central concept is ‘flow’, which signifies health and prosperity (e.g. the undisrupted flow of blood and air through the body). When flow is blocked, this is considered shameful and humiliating. During the Rwandan genocide, perpetrators made a display of abusing and killing their victims in ways that would block flow, such as cutting their victims’ Achilles tendons, cutting off their genitals, or impaling them through the digestive tract (Hagengimana & Hinton, 2009). The purpose of blocking flow was not only to humiliate victims and induce terror, but specifically to burden survivors with memories that triggered humiliation and anguish, known as ‘humiliation memories’ (Hinton & Good, 2016). Humiliation memories may also revolve around the way a loved one’s body is handled after death. During the International Criminal Court trial against Ongwen (a Ugandan former child soldier suspected of atrocity crimes), expert witnesses testified about the trauma caused by witnesses’ inability to bury their loved ones according to Acholi rituals (Schot, 2021). For example, one Acholi ritual is to strangle a goat and eat it several days after the burial, so the the spirit of the deceased can flow through the goat into the people who eat it.

Cultural variations in the experience of negative life events have important implications for investigative interviewers. Events that may sound traumatic to the interviewer, such as surviving a war, may not have lasting negative consequences for the interviewee (e.g. if their culture places a high value on accepting one’s fate and enduring suffering; Friedman et al., 2004). Conversely, events that may not sound traumatic to the interviewer, such as being unable to strangle a goat and eat it shortly after a burial, may be experienced as highly traumatic by the interviewee (cf. Schot, 2021). Such mismatched cultural expectations could result in misunderstandings and misinterpretations in the interview, as we consider in more detail later in this article.

The expression of trauma

Theoretical models of PTSD have predominantly been developed based on people from Western, Educated, Industrialised, Rich, and Democratic (WEIRD) societies (Henrich et al., 2010), and are consequently likely to emphasise Western cultural norms, beliefs, and values (Jobson, 2009). Indeed, PTSD tends to be more prevalent in WEIRD societies (e.g. reported rates as high as 3.5% in United States samples) than in non-WEIRD societies (e.g. rates of 0% in a Yoruba-speaking Nigerian sample and 0.3% in a Chinese sample; Hinton & Lewis-Fernández, 2011). Cross-cultural differences in PTSD prevalence rates may partly be explained by how negative life events are experienced, but also partly by how ‘trauma’ is articulated and non-verbally expressed in an interview (e.g. Hinton & Lewis-Fernández, 2011; Marsella & Yamada, 2010). A problem with using diagnostic descriptions such as PTSD is that patterns of symptoms associated with post-traumatic responses have been found to vary so greatly across time, place, and social subgroup, that no clear PTSD construct can be applied universally (Hinton & Lewis-Fernández, 2011). Western accounts of why people continue to suffer after the event assume that the lingering sense of threat is perpetuated by avoidance of reminders of the event and negative appraisals of what happened to them (e.g. Ehlers & Clark, 2000). Yet, avoidance is not always a typical response (Hinton & Lewis-Fernández, 2011). Similarly, negative appraisals –such as greater mental defeat, fewer control strategies, and more permanent changes in sense of self– are typically associated with PTSD in individualistic societies but not in collectivistic societies (Jobson & O'Kearney, 2008).¹ Alarmingly, mental health disorders tend to increase among forced migrants the longer they have lived in Western host countries, including increases in comorbidity of PTSD with depression or anxiety (Nickerson et al., 2017). A potential contributor to this negative development could be a lack of understanding by mental health professionals and others in the migrant’s new cultural environment.

Whereas the focus in Western cultures tends to be on psychological symptoms, somatic expressions of trauma are common in many non-Western groups. Examples include dizziness and neck soreness in Cambodian refugees, a sense of bodily heat in Senegalese and Salvadorian refugees (Hinton & Lewis-Fernández, 2011), and pain in the stomach and lower back in South African survivors of violent crime (Weiss et al., 2023). The embodied response to trauma reported by survivors of the Rwandan genocide, a cultural syndrome named ihahamuka, is characterised by reported shortness of breath and blocked flow (Hagengimana & Hinton, 2009). Many physical symptoms do not even have a direct translation into English, such as khyâl (a wind-like substance that must not be interrupted) in Cambodian refugees, ñakary (a sense of collective suffering due to external factors) in Peruvian survivors of guerrilla violence, and Han (a sense of unresolved pain caused by unacknowledged injustice) in Korean survivors of sexual enslavement by Japanese soldiers (Bourke, 2022).

If investigative interviewers, legal practitioners, and mental health professionals draw on their own cultural expectations of how a traumatised person will present, they may not recognise trauma in someone who does not conform to their norms. An illustrative example is the discussion of Ongwen’s mental state at the International Criminal Court (cf. Nistor et al., 2020). A Western psychiatrist concluded that Ongwen was probably not depressed because his tone and the jokes he made suggested that he was in a good mood. Yet, the psychiatrist was unaware that in the tonal Acholi language, tone conveys the meaning of the word. Further, he probably mistook a serious comment about wanting to eat white ants, for a joke.² Similarly, asylum seekers and people from indigenous cultures are sometimes perceived as not sufficiently emotional or not displaying the ‘appropriate’ emotion when they are talking about negative life events (Herlihy et al., 2010; Porter & ten Brinke, 2009). Such ‘inappropriate’ emotional displays can be interpreted as a sign of deception. For example, rape survivors are perceived as less credible if they display neutral or ‘incongruent’ emotions (Kaufmann et al., 2003), even though this may reflect a coping mechanism or cultural expectations.

In most of the academic literature on the effect of emotional displays on perceived credibility, cultural differences have been overlooked. In a review of victim credibility in child assault cases, Voogt et al. (2019) note that over 80% of studies have been conducted in the United States, with a large proportion using student samples. Similarly, Van Doorn and Koster (2019) included only studies conducted in Western cultures in their review of the literature on emotional victims and their impact on credibility. Yet, there is clear cultural variability in emotion recognition. In a meta-analytic review, Elfenbein and Ambady (2002) conclude that emotions are more accurately understood when judged by members of the same national, ethnic, or regional group. This own-group advantage may be a result of subtle stylistic differences across groups in the expression of emotions, but the evidence is mixed (see Boucher & Carlson, 1980). According to Elfenbein and Ambady, cultural variability in accuracy of emotion recognition may also reflect translational differences (Mesquita & Frijda, 1992). This idea is supported by another meta-analysis showing that emotion recognition based on speech patterns and non-linguistic vocalisations is better when judging members of one’s own cultural group (Laukka & Elfenbein, 2021). Even in monocultural settings, investigators are not good at assessing credibility based on facial or vocal expressions of emotion (Bond & DePaulo, 2006), and the above-mentioned findings suggest that this is particularly problematic in cross-cultural settings.

Memory for negative life events

Negative life events are not only experienced differently across cultures, but also remembered and reported in different ways. In investigations, interviewees need to draw on autobiographical memory to report about witnessed events. However, autobiographical memory is fallible and fluid and may be influenced by trauma, culture, and interactions between the two (Given-Wilson et al., 2018). The cultural context and immediate social circle in which an individual grew up will influence what is remembered and talked about, so memory reports will reflect culturally and personally defined scripts. These scripts are further affected by contextual adversity; those who have been exposed to ongoing negative life events typically present with less specific or ‘overgeneral’ memories (Moore & Zoellner, 2007). Moore and Zoellner conclude that the primary mechanism behind this overgenerality is not exposure to negative life events per se, but rather the common occurrence of mental health disorders such as PTSD and depression in people who have experienced negative life events, as well as their frequent comorbidity (see also Graham et al., 2014). In this section we consider the literature on cultural differences in emotionality and specificity of autobiographical memory, and how these may interact with negative life experiences.

In studies comparing American (individualistic) and Chinese (collectivistic) adult participants, Wang and colleagues found that Americans’ autobiographical memories focused more on the self (e.g. success, frustrations, dreams and nightmares), while Chinese participants’ memories focused more on relationships and context (e.g. family activities, playing with neighbours, teachers at school; Wang, 2001; Wang & Conway, 2004). Further, they found that Americans were more likely to spontaneously mention emotions in their descriptions and rated their memories as more emotional (both positive and negative) than Chinese participants. American memory reports were also much longer (sometimes more than twice as long) and made more references to a specific episode, such as getting stung by a bee, versus general descriptors of routine events, such as the daily journey to school.

In an eyewitness memory paradigm, Anakwah et al. (2020) found that mock witnesses from Western European (individualistic) cultures reported more details than mock witnesses from sub-Saharan African (collectivistic) cultures, especially when the event was witnessed in a non-native setting. Similar findings have been reported in other individualistic-collectivistic comparisons, including Americans versus Israelis (Gur-Yaish & Wang, 2006), Caucasian Americans versus Asian Americans (Wang & Ross, 2005), and Australians versus East Asians (Jobson & O'Kearney, 2008). Independent self-construal, typical of individualistic societies, promotes greater accessibility to autobiographical memories and facilitates reporting of more specific details (Hope et al., 2022; Humphries & Jobson, 2012; Ross & Wang, 2010; Wang, 2021). In sum, people with independent self-construal tend to provide more self-focused, emotional, detailed, and specific memory reports. In contrast, people with interdependent self-construal typically provide more information about the bigger picture, including contextual information and relationships between people.

When interviewing people about negative life events, cultural background can interact with trauma to influence how a memory is reported (Jobson, 2009, 2011; Jobson & O'Kearney, 2006). For example, Jobson and O'Kearney (2006) found that Australians’ and East Asians’ memory descriptions of everyday events were about equally specific, but Australians’ descriptions of negative life events were significantly more specific than East Asians’ descriptions. Further, like Americans in Wang et al.’s studies (e.g. Wang, 2001; Wang & Conway, 2004), Australians were more likely to choose an everyday event that focused on the self (e.g. academic achievements) and described it with more reference to personal needs, dislikes, achievements and opinions (referred to as ‘autonomous orientation’) than East Asians. However, these cultural differences in focus and autonomous orientation were not observed for memories of negative life events. Thus, general research on cultural differences in memory descriptions may not fully replicate for memories of negative life events. In a follow-up study, Jobson (2011) found that PTSD was associated with fewer references to personal needs, dislikes, achievements and opinions for people from individualistic cultures, but more such references for people from collectivistic cultures. Jobson (2009) proposes that cultural influence should be included in an explanatory model of trauma.

In investigative interviewing contexts, cultural differences and negative life experiences often go hand in hand. Asylum seekers and refugees, who typically come from collectivistic societies and often experience post-traumatic stress, are consistently found to provide less specific descriptions (Graham et al., 2014; Khan & Haque, 2022) with more discrepancies between statements (Herlihy et al., 2002). Because a detailed account tends to be considered more credible and useful, a lack of specificity may disadvantage an interviewee (e.g. Herlihy et al., 2010; Herlihy et al., 2012). It could also harm criminal investigations, since details specific to an incident and time frame are needed to pursue a criminal charge (Connolly & Read, 2006).

Barriers to disclosure

In some cultures, disclosure in an investigative interview could be socially, emotionally, physically, or economically more damaging than silence. One potential barrier to disclosure arises from the cultural concept of power distance, the relationship between individuals varying in power (Hofstede, 1983). In societies with high power distance, more deference is given to authority figures than in societies with low power distance. There are vast differences in power distance between societies (e.g. power distance tends to be relatively low in most Western cultures), but also within societies (e.g. even in Western societies, high power distance can be observed within specific groups, such as religious communities). If power distance is high and the abuser is an authority figure, interviewees may find it shameful to betray that authority figure’s trust despite the abuse (Bourke, 2022; Gyulai et al., 2013; Katz et al., 2022; Weiss, 2002; Xie et al., 2017). For example, in the high power distance culture of Indonesia, rules of behaviour to show politeness and proper manners known as tata karma may prevent children from disclosing abuse to police investigators (Hope et al., 2022). Gul and Schuster (2020) note that a culture of honour may encourage tolerant judgments of men’s sexual aggression against women, especially in situations when the sexual assault occurs within an intimate relationship (Vandello & Cohen, 2003, 2008) and when the victim’s pre-rape behaviour is perceived as threatening to the perpetrator’s reputation (Canto et al., 2017). A victim or witness who fears harm to themselves or family members resulting from disclosure, may be reluctant to talk.

Shame also poses a significant barrier to disclosure in various investigative contexts. Asylum seekers interviewed by immigration officials about the sexual violence they had suffered, cited shame as one of the main reasons for finding it difficult to disclose personal details (Bögner et al., 2007). Shame, humiliation, stigma, and isolation can similarly explain the silence of male victims of violence, as seen in a case study of married men in Kenya who were subjected to domestic abuse (Gitonga, 2021). Of course, shame and stigma may also explain why men in Western cultures are less likely to report being victims of intimate partner violence (Taylor et al., 2022). Bourke (2022) argues that there are many contexts in which silence or rites of forgetting may preserve a victim’s dignity by preventing shame and stigma, ranging from nineteenth-century Mexico to late twentieth-century Rwanda.

Uchida et al. (2020) discuss how interdependency created by the social and economic environment can inhibit disclosure. The success of collective activities that are crucial in many rural societies, such as rice-cropping communities in Japan, relies on mutual trust, which could be damaged if community members turn on each other. In Japan, ‘gaman’, the trait of perseverance and patience when faced with adversity, is socialised from an early age (Littler, 2019). Similarly, in many Arabic-speaking communities, families play a central role in life decisions. Because the honour and needs of the family are prioritised over individual needs, individuals may be reluctant to make disclosures in investigative interviews (Al-Krenawi et al., 2009; Beaini & Shepherd, 2022).

Cultural taboos on language pose further barriers to disclosure. In many cultures, people consider it socially unacceptable to describe sexual acts and therefore use euphemisms or vague language instead (see e.g. Hope et al., 2022; Katz et al., 2022; Xie et al., 2017). In South African isiXhosa and Southern Sotho cultures, there is an entire language code to discuss sexual matters in a respectful manner, called hlonipha, which involves vague language, politeness, and avoiding profanities (Thetela, 2002). Similarly, Rwandan witnesses in the Seromba case used the term ‘forced marriage’ to refer to rape, and judges at the Special Court for Sierra Leone referred to sex as ‘having time with women’ (Combs, 2010). Such cultural norms can hamper investigations in various ways. First, interviewees may be unable or unwilling to describe sex crimes to the degree of detail required by investigators. Second, certain cultural expressions may imply consent when there was none. Finally, interviewees may be so embarrassed by the interviewer’s questions that it hampers rapport in the interview. This factor has been largely overlooked in guidelines for investigative interviewers. We reflect upon potential ways to improve this in the section on Guidance for cross-cultural interviewers.

Cross-cultural communication

Effective communication in investigative interviews is paramount to obtaining complete and accurate accounts from interviewees. However, in societies with high power distance, the desire to agree with a person in authority may outweigh the desire to provide an accurate description. An illustrative example is provided by De Bruïne et al. (2018), who found that Sub-Saharan African respondents were significantly more likely to respond ‘yes’ than Western European respondents when an experimenter (likely perceived as an authority figure) asked them if they had seen an object before, regardless of whether they had actually seen it. This tendency is known as ‘acquiescence response style’ (Cheung & Rensvold, 2000). In investigative interviews, this means that interviewees from high power distance cultures may be more prone to reporting what they believe the interviewer wants to hear, rather than what they remember.

Cross-cultural investigative interviews are also affected by the distinction between high– and low-context communication (Hall, 1976). High-context communication, commonly observed in collectivistic societies, is indirect, focuses on context, and places high value on relationships. Low-context communication, commonly observed in individualistic societies, is direct, focuses on content, and places high value on logic and deductive thinking (Gelfand & Dyer, 2000). Research on Dutch police interviews with suspects from low-context (Dutch) and high-context (Moroccan) cultures showed that rational arguments were more effective in getting case-related personal information from Dutch interviewees than from Moroccan interviewees (Beune et al., 2010; see also Beune et al., 2009). Interviewers should adapt their interviewing style, questions, and strategies to the type of communication that allows the interviewee to express themselves best (see also Taylor, 2018). We provide more specific recommendations on how to achieve this in the section on Guidance for cross-cultural interviewers.

Finally, communication between the interviewer and interviewee may be hampered by the way in which hearsay information is reported. In some cultures, it is customary for people to describe what their family member or close friend has told them as if they had seen it themselves (i.e. vicarious memories; Pillemer et al., 2015). This has been observed many times in the context of international criminal trials (e.g. Combs, 2010; O'Brien & Kebbell, 2014; Schot, 2021). In close-knit communities that have experienced significant negative life events together, such as war crimes or natural disasters, individuals’ memories often merge into an integrated communal understanding of what happened (Cohen, 2012; Schot, 2021), a ‘collective memory’ (Hirst & Echterhoff, 2008). Collective memories in turn reinforce cultural norms relating to one’s family and behaviours that are shared, and shape the collective identity of the community (Uchida et al., 2020). In legal contexts, this means that a witness might testify about hearsay evidence without identifying it as such. If investigative interviewers are aware of this, they can make careful attempts to establish the source of the information by asking follow-up questions, even if the interviewee initially describes it as their own memory. In some cases, this will elucidate who actually observed the event. For example, a witness at the International Criminal Tribunal for Rwanda initially testified that he himself had witnessed a massacre and went on to describe the entire event in detail, but after questions from the defence counsel it became clear that he had run away as soon as the first shots were fired and had later heard what happened from his family members (Combs, 2010). In other cases, even in-depth questioning cannot elucidate the source of reported information. This can be explained by the cognitive process of source confusion (Johnson et al., 1993), where a person genuinely believes that they observed something themselves even though the information came from an external source (or vice versa).

Interpreters in investigative interviews

Investigative interviews in cross-cultural contexts often require an interpreter to translate between the interviewer and interviewee. Walsh et al. (2020) conducted a survey with 66 investigators and 40 interpreters from the United Kingdom about their beliefs on interpreter-assisted interviews in the context of international criminal investigations. They found a striking lack of consensus within and between the two groups on (1) the extent to which they collaborated in planning the interview, (2) the interpreter’s role in establishing rapport, (3) the accuracy of the interpretation, and (4) the extent to which the interpreter’s interventions were disruptive. For example, most interpreters (68%) indicated that they ‘always’ helped to build rapport, while most investigators (62%) indicated that interpreters ‘sometimes’ helped to build rapport. Elaborating on their answers, most interpreters explained that they helped to establish trust, put the witness at ease and considered themselves to be a vital link between investigator and interviewee, but some other interpreters felt that it was not their role to build rapport and that this could compromise neutrality. Similarly, whereas many investigators commented that interpreters’ knowledge of cultural issues was important in establishing rapport, others noted that it was not the interpreter’s role or could make the interviewee feel uncomfortable.

In another study on investigators’ experiences with interpreter-mediated interviews, Goodman-Delahunty and Howes (2019) conducted structured interviews with 121 experienced interviewers from policing, military, and intelligence organisations in Australia, Indonesia, the Philippines, Sri Lanka and South Korea. Again, there was considerable variability in how interpreters were involved in rapport-building. Investigators discussed positive effects of interpreters, such as facilitating communication and rapport, as well as negative effects, such as disrupting the flow, increasing the length of the interview, and increasing the complexity of the social dynamics. According to the authors, many investigators erroneously believed that word-for-word translation would resolve communication challenges, even though professional codes recommend equivalence of meaning rather than equivalence of words.

An example from a murder case described by Filipović (2007, 2013) illustrates the complexities associated with translating for equivalence of meaning. In this case, the suspect had carried an unconscious woman down the stairs and dropped her, after which she died. The crucial question was whether the suspect had dropped her accidentally or intentionally. To distinguish between these two, the investigator asked him whether ‘she fell’ or he ‘dropped her’, using ‘drop’ in its intentional sense. The interpreter translated this to ‘se cayó’ (‘she fell’) or ‘la botó’ (‘threw her’). The suspect answered ‘se me cayó’ (‘to me it happened that she fell’), which clearly indicates a non-intentional action in Spanish. However, the interpreter translated his answer to the ambiguous ‘I dropped her’, which the investigator had used in his question in its intentional sense. Thus, due to subtle translational issues, it seemed like the suspect confessed to intentionally dropping the woman down the stairs, even though he was saying exactly the opposite.

Challenges with interpreters have also been observed in asylum interviews. A systematic literature review revealed three potential problems related to interpreter-assisted interviews in asylum cases (Selim et al., 2022). First, interviewees may mistrust the interpreter and worry about breaches in confidentiality, particularly if they believe that the interpreter shares the same values as their persecutors. Second, the interpreter may lack specific cultural knowledge (e.g. about the interviewee’s religion) to translate statements accurately, which could lead to the interviewer mistaking the interpreter’s lack of knowledge for a lack of expected knowledge on the part of the interviewee, with potentially severe consequences for their asylum claim. Third, the interpreter could unintentionally distort the tone of the interviewer’s questions, for instance making a clarification question sound like a challenge, which could damage rapport between the interviewer and interviewee.

The use of interpreters also influences disclosure in investigative interviews. A review of empirical research on interpreter influence in investigative interviews revealed that fewer details are reported in interpreter-assisted interviews than in interviews conducted in the interviewee’s native language (Evans et al., 2019). Evans and colleagues propose two potential explanations: (a) the interviewee may hold back information to prevent overloading the interpreter, and (b) the interpreter may fail to translate everything. This makes sense as interpreting poses high demands on cognitive resources, hence limitations in working memory and forgetting may account for incomplete translations. In addition, as discussed previously, a lack of trust may inhibit disclosure. Further, in interviews about negative life events, emotions evoked in the interpreter could hamper their ability to render a neutral and complete translation (cf. Morrison et al., 2022; Wilson & Walsh, 2019). Finally, cultural biases and lack of specialised knowledge can result in interpreters editing an interviewee’s statements or replacing a specific term with a generic one. Asylum interviews based on fear of religious persecution provide a good example, particularly if features of the religion, forces, or spiritual entities are not part of societally recognised knowledge (Selim et al., 2022).

To circumvent the above-discussed problems with interpreters, investigators could decide to interview individuals who speak multiple languages in their second language instead. However, interviewees may have difficulty describing emotionally laden memories in their second language. Research shows that emotions are expressed differently in a person’s first and second language. For example, Schwanberg (2010) asked Spanish-American PTSD patients to describe a traumatic childhood experience and rate the intensity of that memory and their PTSD symptoms, first in their first language (Spanish) and then in their second language (English), or vice versa. Regardless of language order, patients who spoke in their first language rated the traumatic memory and their PTSD symptoms as significantly more intense. Thus, descriptions of negative life events in a second language may sound less emotional than descriptions in the first language. This can work against interviewees if an interviewer is looking for credibility cues (Herlihy et al., 2010; Kaufmann et al., 2003; Porter & ten Brinke, 2009).

In some cases, interpreters may not seem necessary because the interviewee is fully bilingual. Yet, even for bilingual speakers, the language in which they report about experienced events can affect the content of their memories and how an interviewee comes across in their presentation. For example, Wang et al. (2010) found that when Chinese-English bilingual participants from Hong Kong were interviewed in English, they provided more elaborate and self-focused memory reports than when they were interviewed in Chinese. The authors suggest that these individuals have different ways of thinking and communicating in different languages. Similarly, bicultural Russian-English bilinguals produced more individualistic narratives when speaking in English than Russian (Marian & Kaushanskaya, 2004). These nuanced findings also illustrate the limitations of relying on the simple binary distinction between individualistic and collectivistic cultural backgrounds to explain differences in behaviour.

In sum, communicating across languages and using interpreters poses additional communicative challenges and barriers to disclosure in cross-cultural interviews about negative life events. Besides translational issues, investigative interviewers should be aware of possible mistrust between the interpreter and the interviewee and the effect this may have on rapport and disclosure in interviews about potentially painful topics (cf. Selim et al., 2022). This problem could be addressed by ensuring there is an effective rapport-building stage with the interpreter, as we discuss in the next section.

Guidance for cross-cultural interviewers

We now turn to the question how the above-discussed research findings may translate into concrete recommendations for investigative interviewers. At first glance, this seems an impossible task. An endless number of cultures and subcultures exist in the world, with endless differences in how individuals express themselves. How can an interviewer possibly know all the ins and outs of the culture of each interviewee and adapt their interviewing style and questions accordingly? That is indeed impossible. The only solution we see is to be aware of potential cultural differences, familiarise oneself with the interviewee’s cultural background if possible, abandon one’s own assumptions as much as possible, and conduct the interview in line with a research-based investigative model. The United Nations has recently adopted the Méndez Principles, replacing coercive investigative interview methods with rapport-based interviews (Méndez et al., 2021). We review below how research findings concerning culture, trauma, and memory could be incorporated into the six Méndez Principles.

The first principle, on foundations, states that ‘effective interviewing is instructed by science, law and ethics’ (Méndez et al., 2021, p. 6). The body of knowledge that attests to the challenges of interviewing across cultures should be incorporated into this principle. Specific examples of this can be provided for each of the three stages identified in the second principle: before, during, and after the interview. This second principle, on practice, highlights that ‘effective interviewing is a comprehensive process for gathering accurate and reliable information while implementing associated legal safeguards’ (p. 15). It specifies steps to be taken before the interview, such as adequately informing the interviewee about the interview purpose and providing access to a lawyer. It also includes that interpreters must be provided ‘for all interviewees who do not speak or understand the language used by the authorities’ (p. 19). As discussed above, this should also be a consideration when an interviewee speaks multiple languages and needs to discuss emotionally laden memories, which are best communicated in the native language (e.g. Schwanberg, 2010). Further, the position of the interpreter should be carefully considered, both in terms of their physical placement in the room (e.g. Goodman-Delahunty & Howes, 2019; Houston et al., 2017; Walsh et al., 2020) and their relationships with the interviewee and interviewer, who should both trust the interpreter to provide accurate and unbiased translations (Evans et al., 2019; Selim et al., 2022).

The second principle also specifies steps to be taken during the interview, such as establishing rapport and using appropriate information-gathering techniques. At this stage, interviewers should consider how cultural differences may affect rapport and disclosure of information. For instance, the interviewer may investigate prior to the interview whether high- or low-context communication is most likely to be effective in the culture from which the interviewee originates (Beune et al., 2009; 2010). Further, the interviewer should be familiar with customary greetings and show cultural awareness in body language, such as interpersonal distance and eye contact. The interviewer can also consider how they can make the interviewee feel as safe as possible to talk about culturally sensitive topics (e.g. Thetela, 2002). For example, it may be useful to discuss at the start of the interview which words the interviewee would be most comfortable using to refer to certain sexual acts. Finally, in concluding the interview, interviewers should be aware that ‘saving face’ is crucial in some cultures, which could interfere with reaching an agreement on the statement to be written up about the interview or the correction of mistakes in such a statement (see Taylor, 2018).

The third Méndez Principle, on vulnerabilities, notes that ‘effective interviewing requires identifying and addressing the needs of interviewees in situations of vulnerability’ (p. 28). This includes interviewees who have specific mental health issues that could be exacerbated by the interview process, for example for asylum seekers (Schock et al., 2015). In this respect, an awareness of cross-cultural differences in idioms of distress and trauma would be beneficial, as well as knowledge of cultural taboos and sensitivities. For example, an interviewee’s complaints about stomach aches or dizziness could point to a culturally specific expression of trauma (e.g. Hinton & Lewis-Fernández, 2011; Weiss et al., 2023). If an interviewee’s vulnerability is not recognised, the interviewer may fail to take appropriate protective measures.

The fourth principle, on training, considers that ‘effective interviewing is a professional undertaking that requires specific training’ (p. 32). An important component of such training should be to raise awareness of cultural factors at each stage of the interview, as illustrated by the examples discussed above. Further, training should address specific cultural assumptions that investigators may hold, including widespread beliefs that inconsistencies, vague descriptions, lack of emotionality, or certain non-verbal behaviours such as gaze aversion are indicative of deception (e.g. Herlihy et al., 2010; Porter & ten Brinke, 2009; Skrifvars et al., 2021). This is particularly important when the interview is about a negative life event and the interviewee is from a different cultural background (Jobson & O'Kearney, 2006). When it comes to an eyewitness, victim, or asylum interview, the expectation that only a coherent and detailed account is indicative of a truthful event may result in an interviewee not being believed or being denied asylum. This is likely to increase despair and shame even more (Bögner et al., 2007; Chantler, 2012) and reduce the efficacy of investigative interviews.

The fifth Méndez Principle, on accountability, states that ‘effective interviewing requires transparent and accountable institutions’ (p. 35) and the sixth, on implementation, stresses that ‘the implementation of effective interviewing requires robust national measures’ (p. 40). Responsibility at the organisational level must be taken to mandate and monitor training that follows these principles, with mechanisms such as course accreditation, supervisory evaluations and ongoing professional development training in place to ensure a high standard. Several countries have already adopted nationwide protocols (see Miller et al., 2018, for an international comparison), of which the PEACE-model in England and Wales is probably the most well-known (for recent overviews, see Bull, 2019; Bull & Griffiths, 2019). Even in those countries, however, the challenge remains to achieve a delicate balance between standardising nationwide interview training and adapting interview protocols to interviewees from different backgrounds. To strike this balance, an approach such as the Standard Interview Method proposed by Powell and Brubacher (2020) may be considered, which includes the standard research-based recommendations on effective interviewing, but also leaves room for tailoring interview guidelines to a specific cultural group or type of interview. For example, collaborations with stakeholders revealed that in interviews with Australian Aboriginal interviewees, linguistic constructions involving the term ‘if’ should be avoided because they are often misunderstood (Hamilton et al., 2016; Hope et al., 2022). At a national level, the findings of a scoping review of policies, barriers and enablers (Pollard & Howard, 2021) may assist with the implementation of guidance tailored for interviewees from diverse cultural backgrounds seeking justice. More work is needed to achieve this.

Conclusion

The findings discussed in this article show that negative life events are experienced and expressed differently in different cultures, and that this influences how events are remembered and disclosed. This can create barriers to communication between the interviewer and interviewee, and potentially the interpreter. These problems can be addressed and potentially avoided by incorporating insights from cross-cultural, legal, and clinical psychology into guidelines and training for all those involved in investigative interviewing and decision-making. Training should involve developing skills for establishing rapport and trust with individuals from different cultural backgrounds and meeting their specific needs, such as a preference for a gender-matched interviewer and the language in which the interview should be conducted. It should also address the assumptions officials may hold about how people remember and narrate life events.

The literature on interactions between culture, trauma, and memory is sparse, and has predominantly focused on broad distinctions between individualistic and collectivistic cultures. We have illustrated how a more comprehensive and nuanced approach to cultural differences may improve information-gathering about negative life events, and how we can incorporate the insights into interview guidelines. Even so, there will never be a complete and exhaustive knowledge base on how culture influences investigative interviews. This means that we need to design procedures that are consistent with research-based interview guidelines while allowing for sufficient flexibility to adapt to the needs of different cultural groups and interview contexts (cf. Powell & Brubacher, 2020). There is no single recipe for effective cross-cultural investigative interviews, but we argue that adherence to the Méndez Principles with added awareness of cultural differences can improve investigative interviewing in a meaningful way.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement n° 802080) and from Unbound Philanthropy.

Notes

1 However, Jobson and O'Kearney (2009) did not replicate these findings, indicating that we need further understanding of the nuanced ways in which people make sense of negative events across cultures.

2 The supposed joke and cheerfulness even led to the comment: ‘Be fair to say that those aren’t the typical symptoms of a man suffering from major depressive disorder, wouldn’t it?’; ICC, The Prosecutor v Ongwen, ICC-02/04-01/15, Transcript T-249, 19 November 2019, p. 51.