Sequential Interviews with Preschool Children in Norwegian Barnahus

This chapter presents and discusses a new sequential interview model that is tailored to the needs of preschool children aged about 3 to 6 years old when they participate in investigative interviews in Barnahus. The extended forensic interview protocol (EFI) from the USA has served as the foundation for the development of the Norwegian sequential interview model (SI). This new model is based on close collaboration between police interviewers and Barnahus staff. The model differs from the standard procedures of investigative interviews in Norway and these differences are highlighted in this chapter. This chapter concludes with suggestions for further research on investigative interviews with preschool children.

consistent with international principles for scientific investigative interviews (Davies and Westcott 1999;Fisher and Geiselman 1992;Lamb 1994;Milne and Bull 1999). The differences between DCM and SI are highlighted in the chapter. The chapter concludes with suggestions for further research into investigative interviews with preschool children.
Two different evaluations of Norwegian legislation on child investigative interview have concluded that the legislation was not adapted to the needs of children when they testify about their experiences of violence and sexual abuse in investigative interviews (Ministry of Justice 2004. The evaluations pointed out the need for more child-friendly interviewing methods. It was particularly noted that preschool children may need more than one single interview to talk about their experiences. Regular investigative interview practice with children in Norway normally lasts about 1 h allowing only one short break during the interview. Against this background, changes were made to the Norwegian criminal code. The new amendments allowed the possibility of adjusting the investigative interview to the individual child's developmental and psychosocial needs during the interview, to a greater degree. The amendments expanded the opportunity for supplementary interviews. The law does not specify which methods must be used in child investigative interviews; however, other than that the interviews shall be conducted in accordance with currently approved interview methods taught at the Norwegian Police University College (Ministry of Justice 2014, see also Chap. 5 in this book).
The lack of satisfactory arrangements to meet the needs of preschool children in the interviews led to the development of the Extended investigative interview project. The project involved participants from the National Criminal Investigation Service (NCIS/KRIPOS), Barnahus and the police in the city of Bergen and Norwegian Centre for Violence and Traumatic Stress Studies (NKVTS). The authors of this chapter were members of the project group: Tone Davik from NCIS as the project leader and Åse Langballe from NKVTS as a researcher.
The aim of the extended investigative interview project was to combine practical experience with scientific research to achieve a more "tailored" investigative interviewing approach for preschool children in Norwegian Barnahus.
Below, we describe and reflect on the new procedures for interviewing preschool children in Norway that resulted from the project. First, we will summarise some of the most commonly described challenges in interviews with the youngest children. We will then describe the main elements of the interview method commonly used when older children are interviewed by the police in Norway [the dialogical communication method (DCM) as described by Gamst and Langballe (2004)] and present the extended forensic interview (EFI) (Carnes et al. 1999(Carnes et al. , 2001, developed in the USA. Finally, we describe how the extended investigative interview project combined these two interview models in new procedures for interviewing preschool children as the SI method.

Challenges Encountered When Interviewing Preschool Children
There are important factors to assess when preschool children are interviewed in investigative interviews, including their language development, memory capacity, vulnerability to suggestibility and psychosocial aspects (Bruck et al. 2006, Lamb andSim 2013). Research has shown that interviewers often fail to ask age-appropriate and understandable questions to children and that this may lead to incomplete information from them (Saywitz et al. 1990(Saywitz et al. , 1993. It is well documented that the use of open-ended questions increases the chances of eliciting accurate and complete information from a witness (Mamon and Bull 1999;Christianson et al. 1998;Fisher and Geiselman 1992), but we also know that open-ended questions (tell me about your father) are experienced as problematic for preschool children, because they do not relate to a context that can help the child understand the intent of the question (Fivush 2002). Focused and context-related questions are alternatives. Focused questions can be defined as closed (yes and no) questions or choice questions. Several studies show that children's answers to closed questions (are you afraid of your father?) and choice questions (did he hit you one time or more than one time?) often lead to incomplete and incorrect answers (Lamb et al. , 2003Peterson et al. 1999).
How to adequately phrase questions in a legal context while still taking the child's developmental needs in account is challenging. The questions should be related to the child's reality and words that the child understands should be used. To achieve this, thorough preparations are required. This entails getting to know how the child expresses themselves and experiences the surroundings. This also involves making the child feel safe and giving them a feeling of mastery during the interview situation (La Rooy et al. 2009;Leander 2010;Sternberg et al. 1997).
From a witness psychological perspective, potential influence on the child is a central theme when children are asked to retell past events (Ceci and Bruck 1995). Forgetfulness, false memories and external circumstances are interrelated factors that can create unreliable testimony from a witness, and specifically from a child, as a result of developmental conditions (Tetzchner 2001). Memory is prone to be negatively affected by elapsed time, and the time from an event has occurred until it is communicated, especially concerning investigative interviews with preschool children (Ceci and Bruck 1995). This is why it is important to provide the child with the opportunity to disclose the incident as soon as possible after its occurrence. On the other hand, children also need time to remember and to be able to articulate traumatic experiences. If the child is allowed to talk over time and on several different occasions, it may facilitate memory and thus help with the difficulties encountered while remembering. Memories of an incident can be reinforced, not only by the child repeating what took place, but also because an adult can assist in helping to explain what has happened and clear up any misunderstandings (Christianson and Granhag 2008). Multiple interviews can thus be in the best interest of the child and needed in order to inform the case under investigation. This requires, however, that interviews be conducted by competent interviewers who use recommended methods (open-ended questions). Research shows that children underreport abuse rather than report abuse that has not happened (Cederborg et al. 2007;Leander 2010). Supportive adults make it easier for a child to talk about abuse. Such support involves time to establish trust and feelings of safety during the interview situation (Hershkowitz et al. 2006).

The Standard Model in Norway: The Dialogical Communication Method (DCM)
To fully comprehend the procedural differences between the regular interview model in Norway (DCM) (Gamst and Langballe 2004) and the sequential interview model for preschoolers (SI), an overview of the phases in DCM will be presented. In addition to the phases, DCM consists of following areas: the interview setting involving formal and physical factors, verbal and non-verbal communication and topic development. These areas are not described in this presentation. The first author of this chapter is one of the two researchers who developed DCM (Gamst and Langballe 2004).
The first phase in DCM is the preparatory phase. The objective of this phase is to reduce stress in the interview situation and to give-and collect-information in order to make the child and the interviewer as prepared as possible for the interview. Criteria towards obtaining this objective are to provide information about the interview to the child's caregiver, obtain information about the child, collect information about the reported offence and conduct a meeting with the legal participants in advance of the interview. The main objective of the second phase, the rapport phase, is to establish contact with the child and an atmosphere of trust, and try to relax the child. Other objectives are to estimate the cognitive level and emotional state of the child, to introduce dialogue as the form of communication during the interview, to explore the child's understanding of the concepts truth and lies and to obtain personal information from the child. The child should thus be allowed to express themselves personally and talk about personal interests and non-intrusive subjects freely. The interviewer also shows themselves and encourages the child to tell the truth. In the third phase, the preliminary phase, the objective is to motivate the child to talk about their experiences, to reduce the disparity of power between the interviewer and the child and to explain the rules of the conversation. To achieve this, the interviewer clarifies their professional role, generalises the situation, provides the child with an overview and structure of the interview and explains the most important rules of the conversation. The fourth phase is called the introduction to the focused subject. The objective is to introduce the focused theme in a neutral manner. The criteria for achieving this goal are to introduce the case by asking open questions stemming from several hypotheses of the suspected crime. The fifth phase, free narrative, aims to obtain as much spontaneous and coherent information from the child as possible. By practising open-ended questions and active listening, the child will be given the opportunity to express themselves in a free narrative. The sixth phase, the probing phase, aims to obtain detailed, comprehensive and consistent information and illuminate the case as deeply and broadly as possible. The child has the opportunity to go thoroughly into separate themes in the narrative by clarifying and expanding on elements in the story. The seventh phase, closing the interview, aims to support the child and establish a basis for a positive end. The aim is that the child should have a sense of being seen, understood and taken seriously.

Developing the Sequential Interview (SI) Model Extended Forensic Interviews (EFI)
The extended forensic interview protocol for children was developed by the National Child Advocacy Centre in Huntsville, Alabama (Carnes et al. 1999(Carnes et al. , 2001, specifically for the youngest children and children with special needs. This protocol splits the interview into several sessions, thus allowing more time for the interview. This acknowledges that time is needed to establish rapport and allows the interviewer to make adjustments throughout the interview process. The USA has more than 15 years of experience with this method, and it appears that extended assessments disclose severe sexual abuse cases to a greater extent; however, it is stressed that more research on the extended approach is needed (Carnes et al. 1999(Carnes et al. , 2001. The phases in EFI described by this research have many similarities with those in DCM.
The interviewer is responsible for planning and organising the session, allotting the time needed for each session and between sessions.
Normally, only one session is conducted a day. The EFI approach underlines the importance of careful preparation, including the collection of information about the child's language skills, cognitive development, family situation and possible trauma experiences. The method also allows the use of props such as pictures or drawing materials to encourage further verbal explanations from the child when needed.

Methodology
A project group was established in 2010 to apply EFI to preschool children by testing it in Barnahus. The Norwegian model of EFI was called sequential interviews (SI).
The first ten SIs were conducted by three different police interviewers and analysed by the Barnahus in Bergen to explore whether SI elicited more abuse-related information than normally practiced interview methods. The results showed positive changes in eliciting more abuse-related information from very young children in the SI interviews compared to the standard interview model, and that the new model was perceived positively by legal representatives and caregivers (project report after testing sequential interviews with preschoolers 2012) ("Rapport etter utprøving av sekvensielle avhør av førskolebarn 2012"). The same year, a pilot group was appointed, consisting of nine expert child interviewers from NCIS and three police districts, the specialist psychologist from Bergen Barnahus and the researcher who is the first author of this chapter. The nine expert child interviewers received SI method training from the project group as well as lessons in preschool children's developmental psychology and specific challenges encountered when communicating with them. This marked the beginning of a structured collaboration between the Barnahus staff and police interviewers in developing the SI model in Norway.
This project used qualitative methods. The investigative interviewers kept journals after each SI session. Minutes were also recorded from two group meetings running for 2 and 3 days, at two different time points, where representatives from the Barnahus staff and the nine experienced police interviewers were present. Group and open discussions based on observations of actual audio-taped interviews using the SI model were also held.
The authors also conducted semi-structured group interviews of four experienced police interviewers and four experienced counsellors from the Barnahus staff, to gain information about experiences from their collaboration with each other.
An overview of experiences from conducting SI in Norway, concentrating on preparations and the actual sequences in the interview, is presented below.

Key Differences Between the Standard Method (DCM) and Sequential Interviews (SI)
DCM procedures were both simplified and extended in the SI to suit the youngest children. It became clear that the following was needed: better preparation prior to the interview-for the child, the child's caregivers and the police; more time and breaks during the interview to enable the child to relax and feel secure; the use of various props and toys; and spending a considerable amount of time throughout the interview allowing the child real opportunities to disclose their experience(s) of violence or sexual abuse. This project also demonstrated new prospects for interdisciplinary collaboration within the context of Barnahus.

Preparations
SI emphasises the importance of thorough preparation. Information gathered from those who know the child well can assist the Barnahus staff and the police interviewer in adapting the situation to the child's needs. These adaptions could include making detailed plans for how the child is received at the Barnahus including: • Who should greet the child?
• What to say during the initial meeting?
• Which playroom (waiting room) is best suited to making the child feel safe and calm?
Challenges regarding young children's language comprehension and memory functions can be solved by obtaining information about the child and their family, surroundings, daily life and experiences, and are needed for formulating questions suited to the individual child. By making a detailed plan of adequate questions, the risk of asking suggestive questions and thus eliciting incorrect information is reduced. The preparation phase of the SI is therefore extended compared with this phase in the DCM.

Legal Participants
It is important to ensure that all participants involved in the interview have scheduled enough time for the SI (4-5 h). It is necessary to inform those in the judicial and police field about what the SI approach entails. For this reason, the judge, 1 police advocate, defence lawyer 2 and the child's legal representative were provided with written information about the SI, and the challenges of interviewing the youngest children as described in an article written by the authors of this chapter (Davik and Langballe 2013). In this way, the participants had the opportunity to prepare and gain more knowledge about the interview situation.

Collecting Information
If one or both parents are the suspect(s), a person from the kindergarten or from the child welfare services will usually accompany the child to the Barnahus. The parents or the professionals who accompany the child receive advice from the interviewer on how they can explain the reasons for going to the Barnahus to the child, what happens on arrival and who accompanies the child. The child also receives information about the actual interview and what it will be about, which is adapted to the maturity and psychological needs of the individual child.
Due to the nature of violence and abuse against young children, the police investigators usually have little information concerning the reported offence. Investigators therefore often need to gather more information from witnesses who know the child-for example, kindergarten staff. Relevant background knowledge includes how and to whom the child disclosed the incident in question, how the person who obtained the child's story responded to the child and the kinds of questions that were asked when the child disclosed the abuse.
In line with DCM, the investigative interviewer will ask the child's caregivers for information about the child's daily life, specific interests, cognitive, verbal and social development. In SI, the caregivers may also be asked to provide information on significant or recent positive experiences that they believe the child would enjoy talking about. This information can be used in establishing rapport and to explore the child's ability to provide free narratives.

Information Sharing
Two meetings are usually held before the interview. The first meeting is commonly a telephone conference to plan collaboration and the practical conduct of the interview. It involves the police interviewer and a counsellor from the Barnahus. A police investigator, and often the police lawyer and a representative from the child welfare services, normally also attend the meeting.
The second meeting is held immediately prior to the actual interview. Those present at this meeting are the police lawyer (in charge of the interview), a police investigator, the police interviewer, other legal representatives (who follow the interview) and the counsellor from the Barnahus. The meeting discusses the challenges that are to be expected when interviewing young children, specific challenges concerning the particular child and the level of information that can be expected from the child. The interviewer's plan for the interview is also presented and discussed.

The First Session
The first session normally lasts about 20-30 min and includes the second and third phases of DCM. A break before the introductory phase provides the interviewer with the opportunity to focus on important elements in the rapport phase, such as building trust and making the child feel secure. Not having to rush, and knowing that the child's concentration span is limited, helps the interviewer stay calm and focused on the child, and able to observe the child's behaviour, language and reactions to questions, various tasks and topics. During this session, the child should also receive information about the reasons for the interview, preferably in a way that is not an invitation to start talking about the actual case.
Props like a picture book, a puzzle or drawing materials can be used in the first session, to help the child start talking. In DCM, drawing material is usually the only play material used. As in DCM, the child should be informed of the basic rules of communication for the interview, including that the child can say I don't know if there is something they simply do not know, and to correct the interviewer if they say something that may be incorrect. The props mentioned above can also be used to examine the child's knowledge of concepts such as quantity, size, colours and shapes. By talking about neutral matters while using props, the interviewer and the Barnahus counsellor explore the child's cognitive abilities.
In the first session, questions should primarily be open, to see if the child understands open questions. The interviewer will also use more specific questions, such as what, who, where and how, to explore the child's responses. In this session, it is also important to encourage the child to tell the truth. Young children's understandings of abstract concepts like "truth" and "lies" are complicated matters in a legal context. Here, however, the interviewer is only required to inform the child about the rules of telling the truth so that the interview meets legal requirements [Criminal Code § 128 (Straffeprosessloven § 128)] and fulfils the regulations of child investigative interviews §10 [Forskrift om avhør av barn og andre saerlig sårbare fornaermede og vitner (tilrettelagte avhør)].
Normally, the interviewer introduces a recent event the child has experienced, and the child will then be asked to say more about itin order to explore the child's ability to master the free narrative form. If not, information collected in the preparation phase will be used to explore the child's abilities to narrate prior experiences.
The session ends when the child starts to get distracted or once the interviewer has an idea of the child's language and developmental level. In taking a break so early in the process, the child will normally experience the interview as something positive, combined with a feeling of mastery and control, and feel secure and more familiar with the interviewer and the interview situation.

The First Break
The first break usually lasts for 45-60 min. During the break, the child is allowed to relax, play and get something to eat-but too much stimulation should be avoided. The break also provides the interviewer time to receive feedback from the Barnahus counsellor and thus an opportunity to change the direction and strategy for the rest of the interview, if needed. The counsellor and the interviewer can also discuss their common experiences from the first session in detail, sharing their impressions of the child's emotional state, language skills and behaviour in the interview situation.

The Second Session
In the second session, preschool children will tend to have an attention span of approximately 15-20 min. The interviewer can decide whether it is appropriate to take breaks during this session. In this presentation of the SI, we choose to present only three sessions in an interview conducted over only one day. The model is flexible, and the numbers of sessions and the possibility of using several days for the interview must be considered by the interviewer and the police advocate, together with the counsellor at the Barnahus as the interview progresses, bearing in mind the best interest of the child.
The interviewer begins by repeating and explaining the reason for the interview to the child. After this, the interviewer introduces the topic of focus to the child. If the child is unable to talk about the experience(s), they can make some drawings together. This might be a good way of re-establishing contact after the break. The interviewer can also present photographs of the child's home and family, or other people related to the case. Using such props can help direct the preschool child's attention to the topic in focus. It is stressed that photographs are introduced in an open manner, such as: Tell me about this house. Tell me who lives with you in this house. To make themes tangible to the child, the interviewer may draw while talking with the child-for example, the child's house, parents, siblings, pets, etc.
If the child brings up abuse or other relevant topics for the case, the interviewer follows up with questions. The follow-up questions should be open ended, but concrete, and adjusted to the child's developmental level. For example, You told me your father hits when he gets angry. Tell me more about your father hitting (or being angry). If this question is too open, a more focused question might be: Tell me who your father hits when he gets angry. If the questions are too advanced for the child, they might lose attention and thus their motivation to complete the interview. Difficult questions also increase the potential of incorrect or short answers, such as yes/no and I don't know.
As a rule, the interviewer should always follow the child's lead throughout the whole interview by not ignoring any conversational initiatives the child takes.

The Last Break
There is a short break of 5-10 minutes before the interview ends. This break offers legal representatives the opportunity to formulate any remaining questions to the child, to be asked by the interviewer. The interviewer is given the opportunity to discuss the case, if needed, with the counsellor. It is important that this break is not long to avoid the child becoming tired and unfocused.

The Last Session
It is often necessary to ask a few final questions on behalf of the legal representatives towards the end of the interview also allowing an opening for a second interview, in case new questions emerge. This last session is normally quite short, lasting 5-10 min.

Final Reflections
The most significant difference between "practice as usual" in Norway (DCM) and sequential interviews (SI) is first and foremost a new form of interdisciplinary collaboration that lasts throughout the whole interview process. In one of the group interviews, a Barnahus counsellor says: I define us as a team, and we will do this together when there is sequential interviewing. (…) Now we [the Barnahus staff] can contribute with our knowledge, working as a team throughout the whole procedure. Start-up isn't when the actual interview begins, but when it is scheduled.
The interviewees describe a productive collaboration during the preparations for the interview; however, there are challenges in balancing professional roles. In particular, one can question whether the specific child-oriented expertise may bias the investigation, by negatively influencing the potential to obtain objective and unbiased information from the child. The SI approach has given both the Barnahus staff and the police the opportunity to become familiar with each other's professional work. The counsellors describe it as a difficult but meaningful task to weave together the child-oriented and investigative perspectives within the framework and goal of the interview. The police interviewers, through cooperation with the Barnahus staff, describe the knowledge that has been mutually gained, and how this serves as a foundation in the development of a culture for cooperation.
Research has shown the importance of familiarity with a proper methodology for interviewing children, and that support and guidance are needed for this knowledge to be sustained (Lamb et al. 2002;Orbach et al. 2000;Powel et al. 2008). An investigative situation involving a young child will usually entail a high degree of systemic stress. Under such conditions, even an experienced and trained interviewer may not always manage to observe what is happening with the child, and between themselves and the child, and be able to find the best practices and solutions in the current situation. It is important that the interviewer receives feedback and good counsel based on interdisciplinary knowledge and a shared understanding of the tasks at hand. It is essential for the success of the collaborative effort that the main aim of the investigative interview is clearly defined, as are the tasks that this entails.
The informants pointed out that the assistance provided to the SI interviewers may differ greatly. They also described providing counselling for the interviewers as a complex task that requires specific expertise, and pointed out the need for a system to ensure continued expertise over time.
A central element of the SI model is the use of props. We have not, however, examined how the props are used and what effects they have on the children's narratives. In order for the props to be properly used, they should be used in a deliberate and structured manner. For example, preschool children may have difficulties relating pictures to a real object, while photographs and drawings can help children explain things (Cederborg et al. 2009;Hewitt 1999). According to Poole and Dickinson (2013), children aged 5-12 years can divide their attention between voluntarily drawing and talking about past events, but we know little about how younger children and those with cognitive impairments master such tasks (Poole and Dickinson 2013).
Extensive use of breaks can both strengthen the child's endurance and concentration and distract or exhaust the child. We have not systematically examined how the breaks are used and how they affect the children in SI.
Although this project has not been scientifically tested, we believe we have described important topics in assisting investigative interviews of preschool children in Barnahus, by systematising experiences of highly qualified and experienced Barnahus staff and police interviewers. The EFI approach is well proven in the USA, with positive results, as described earlier in this chapter. Because investigative interviews of preschool children are complex and require knowledge from multiple theoretical perspectives, there is a need for a combination of scientific, theoretical knowledge and systematically described experiences.
Our informants described their realisation that very young children, in difficult situations, were faced with unsuitable methods and procedures that simply did not work. They described how working together has paved the way for better interviewing practice. The project forms the basis of further research into the strengths and weaknesses of SI, and how this approach can best be implemented in Barnahus.

Notes
1. Under the former legislation, before October 2, 2015, a judge was the formal leader of the interview. 2. Under the former legislation, the defence lawyer had an expanded opportunity to also be present during the first investigative interview of the child.