Forensic Psychology

by Graham M Davies

  7.2 SHORTCOMINGS AND CONSEQUENCES OF TRADITIONAL INVESTIGATIVE INTERVIEWS

  Conducting an investigative interview is a demanding and difficult task. Police often feel tremendous pressure to solve a case quickly and must consider a host of factors when interviewing witnesses, including their cognitive and social abilities. They may interview witnesses in less than ideal conditions (e.g. with visual and auditory distractions). Witnesses may be under stress or possible coercion from others concerning what to say (or not say), and they may be hesitant to disclose information for fear of embarrassment or punishment to others. Interviewers must listen closely and formulate appropriate follow-up questions while trying to avoid seeming distracted or disengaged. Moreover, in some situations, there may be no additional evidence (e.g. DNA), making the investigative interview crucially important for solving the case, prosecuting the perpetrator, and enhancing public safety. Despite the numerous challenges involved in conducting investigative interviews, interviewers must strive to conduct thorough interviews with each and every witness.

  To aid in this cause, researchers have developed several best-practice interviewing techniques. These techniques were developed in response to shortcomings in investigative interviews that were observed in several high profile daycare child sexual abuse cases in the 1980s and 1990s (see Ceci & Bruck, 1995) and in several field studies (Fisher, Geiselman, & Raymond, 1987; George & Clifford, 1992; Warren, Woodall, Hunt, & Perry, 1996). These shortcomings included the following:

  Interviewers established little rapport with interviewees and frequently interrupted them.

  Interviewers often used complex and compound questions or language (e.g. “So you were in a supine position?”).

  Interviewers used few open-ended prompts (“Tell me what happened”), and instead tended to rely on asking a series of direct and potentially pre-scripted questions (“How old was he?”), with frequent suggestive or leading questions (“Was he fair skinned?”; “He touched you a second time, didn’t he?”).

  Questions were often asked in a predetermined order, and interviewers sometimes asked interviewees to elaborate on information only after the topic had changed.

  Interviewers often neglected to tailor the interviews to interviewees’ developmental, social, and cognitive abilities (e.g. asking complex questions with children), but rather followed a similar interview protocol for all individuals regardless of age, IQ, and/or any disabilities.

  These shortcomings, and the very real consequences that may result when best practices are not followed (see Case Studies 7.1 and 7.2), underscored the need for interviewing techniques that could increase the quantity and accuracy of information that witnesses report. Improving interviewing techniques can help ensure justice for all – protecting victims and innocent suspects alike. Interviewing guidelines and protocols were thus developed with a basis in developmental, cognitive and social psychological theories. Next, we review one such technique – the Cognitive Interview (CI).

  PHOTO 7.1 Police officers may interview witnesses in less than ideal conditions.

  Source: © Image Source/Getty Images

  CASE STUDY 7.1 THE CONSEQUENCES OF POOR QUALITY INVESTIGATIVE INTERVIEWS

  The consequences of interviews that do not adhere to best practice can be profound for all involved, including the alleged victim. In a case in Scotland, a mother alleged that her 6-year-old daughter was abused by the girl's biological father. The mother told investigators that, after a visit from her father, the girl had a “reddish looking vagina”. Without appropriate planning and background investigation, a series of poor-quality, suggestive interviews were conducted that ultimately formed the basis of legal proceedings in which the child and mother testified. However, because the initial investigation was scant, investigators did not fully appreciate that the mother and father were amidst a bitter divorce, and that the mother had possibly planned to make a false allegation of abuse in an effort to secure custody of her child and prevent all contact with the father. The poorly-conducted interviews were played to the court, and the girl was called for live cross-examination. The child was asked more than 200 questions about statements made in her interviews and broke down in tears from mental exhaustion after being directly accused of lying about the allegations. To this day, she still requires ongoing therapy and remains estranged from her father. The social workers who conducted the interviews were severely criticised by the judge in the case. Their employers placed them on “sick leave”, and they later moved on to new careers. Ultimately, we will never know whether abuse occurred, but the absence of properly conducted interviews had serious consequences for the child.

  CASE STUDY 7.2 THE CONSEQUENCES OF FAILING TO RECOGNISE THE VALUE OF HIGH- QUALITY INVESTIGATIVE INTERVIEWS

  Failure to recognise the value of good forensic interviews comes with a cost. In a recent case in England, a 12-year-old girl was interviewed twice by police about allegations of rape by her stepdad. The abuse was suspected by her mother who arrived home and found her partner in the girl's bedroom late one evening. The girl disclosed to her mother that abuse had occurred, and they went to the police who investigated and conducted two best- practice (i.e. ABE) interviews. In the interviews, the girl described three incidents in detail. She described the first time she was raped, the last time, and another time that “stuck out” in her mind because her stepdad was very drunk. The police referred the case on, and it was scheduled to be tried in the High Court. While preparing for the case, the defence attorneys asked a memory expert to review the interviews, and the expert provided a report that concluded that the interviewers followed correct procedures, and thus the testimony would be “robust to challenge”. Therefore, the defence attorneys elected not to file the report from their expert. Instead, they argued that the girl was making up the allegations and telling lies. In court, the prosecuting attorneys did not lead with evidence from the girl's interviews and instead relied on the mother as a key witness, although she had not actually witnessed the abuse herself. The jury was not persuaded and returned a “not guilty” verdict. It is likely that had the attorneys relied on the interviews, a “guilty” verdict would have been obtained.

  7.3 THE COGNITIVE INTERVIEW (CI)

  The CI (Fisher & Geiselman, 1992) is a toolbox of interview techniques used primarily with cooperative adult witnesses, since some of the cognitive components are less suitable for children (see Fisher, Ross, & Cahill, 2010; Fisher, Schreiber Compo, Rivard, & Hirn, 2014; Memon, Meissner, & Fraser, 2010 for reviews). Originally, the CI was derived mostly from cognitive principles, but later came to incorporate many social communication elements, as well. It is used around the world and is particularly popular in the UK.

  7.3.1 Overview of the CI

  First, interviewers establish rapport and outline the ground rules of the interview (e.g. interviewees should do most of the talking; the interviewer should avoid guessing). Second, interviewers ask the interviewees to tell them everything they remember. Third, interviewers instruct the interviewees in the use of different retrieval strategies to aid them in elaborating further on any recalled information (e.g. relaying the event from a different perspective, drawing a sketch). Finally, interviewers review the information that interviewees recalled and provide them with contact information in case they remember any additional details in the future. The techniques used in the CI can be broken down into the following components: social dynamics, cognitive processes, and communication (see Figure 7.1).

  FIGURE 7.1 Components of the Cognitive Interview

  7.3.1.1 Social dynamics

  The component of social dynamics involves the elements of building rapport and active witness participation. Rapport can be established through verbal behaviours (e.g. expressing interest in the interviewee, disclosing information about themselves as interviewers) and non-verbal behaviours (e.g. head nodding, making eye contact). Some researchers have found that rapport increases the accuracy of information that adults report, while decreasing incorr
ect details or misinformation (see Vallano & Schreiber Compo, 2015 for review). Interviewers can encourage active witness participation by conveying that interviewees should lead the interview and do most of the talking; by asking open-ended questions that activate recall memory and require interviewees to elaborate on responses; and by avoiding interruptions.

  7.3.1.2 Cognitive processes

  It is difficult to recall experiences in great detail. In fact, it is a normal feature of human memory that an event account is never complete at the first recall session. Studies show that both children and adults report new correct details with additional interviews (La Rooy, Pipe, & Murray, 2005; Odinot, Memon, La Rooy, & Millen, 2013). Retrieving a memory and reporting it to others is especially taxing if there has been a long delay from the target event to the interview. In the pub fight example at the beginning of the chapter, police conducted interviews immediately. But what if an investigative interview is conducted days, weeks or even years following an alleged event? It is imperative that interviewers are aware of the limitations of human memory and that their interviewing procedures follow suit. Investigative interviewing techniques like the CI seek to minimise these limitations by relying on several key cognitive processes.

  The cognitive processes component of the CI involves five main elements: (1) multiple and varied retrieval, (2) context reinstatement, (3) limited cognitive resources, (4) minimisation of guessing, and (5) minimisation of constructive recall. In light of research evidence demonstrating that witnesses often recall new information in each telling, the CI instructs interviewees to use multiple and varied retrieval or recall event details multiple times and in different ways. For example, this means that interviewers can ask interviewees to recall the event in reverse order or from a different perspective (e.g. the perpetrator’s perspective). Witnesses may also be asked to mentally reinstate the context (i.e. mentally put themselves back to the time and place of the event). These techniques have been shown to increase the amount of information that interviewees recall (Anderson & Pichert, 1978; Smith & Vela, 2001). Because interviewees have limited cognitive resources, the interviewers may also suggest techniques to facilitate concentration, such as closing their eyes, which increases the number of correct, but not incorrect, details recalled (Perfect et al., 2008).

  To minimise guessing, interviewees should be told to respond with “I don’t know” if they do not remember the requested information. Finally, to minimise constructive recall, interviewers should avoid suggestive and leading questions. These types of questions may lead interviewees to provide a particular response (e.g. “He had a knife, correct?”) and may introduce misinformation (i.e. incorrect information) into interviewees’ accounts. Interviewers should also be careful of nonverbal behaviours that may indicate satisfaction with or disapproval of particular responses. For instance, smiling or head nodding may indicate satisfaction with a response, whereas a furrowed brow or perplexed look may indicate disapproval.

  7.3.1.3 Communication

  The final component of the CI is communication, and it involves encouraging elaborate responding and use of non-verbal techniques. Interviewers should emphasise that interviewees report any and all details that they recall, including seemingly insignificant and contradictory information. In other words, interviewees should let others decipher whether the information is pertinent to solving the case. Another retrieval strategy involves asking interviewees to recall information non-verbally, such as through the use of a sketch. Some information can be more easily depicted visually, such as the scene of a crime.

  7.3.2 Empirical Support for the CI

  Numerous field and laboratory studies have been conducted to examine the effectiveness of the CI, and the results are robust and reliable. Field studies can give a sense of what it is like to implement and conduct the CI in the “real world”, whereas experimental studies conducted in the laboratory can help researchers draw causal conclusions about the interviewing techniques. Generally, the CI results in a 25% to 50% increase in the amount of information that witnesses report compared to standard police interviews (Fisher, Milne, & Bull, 2011; Fisher et al., 2010; Fisher et al., 2014). For example, Fisher, Geiselman, and Amador (1989) trained detectives on the CI and found that from pre- to post-training interviews, these detectives elicited 47% more information from witnesses. Also, the detectives trained on the CI elicited 63% more information from witnesses than detectives not trained on the CI.

  Researchers can rely on a technique known as “meta-analysis” to combine the results of many studies, and thus, make broader conclusions about a particular topic. According to one such meta-analysis, which considered the results of 47 published articles on the CI (see Memon et al., 2010), the CI: (1) increases the number of correct details recalled; (2) slightly increases the number of incorrect details recalled (e.g. saying the perpetrator wore a black shirt instead of a blue shirt); and (3) does not increase the number of confabulations or self-generated details (e.g. saying there was a weapon involved when there was no weapon at all). Although there is a potential small increase in incorrect details recalled, this outcome can be avoided if interviewers encourage witnesses to monitor the information they report, by, for example, not guessing and indicating when they do not know the answer.

  7.3.3 Variants of the CI

  Unfortunately, empirical support for a technique does not necessarily mean that it is implemented easily or successfully in the field. A key complaint among law enforcement is that the CI is overly burdensome and takes too long to administer. Some feel that they are insufficiently trained to administer the protocol (Dando, Wilcock, & Milne, 2008; Kebbell, Milne, & Wagstaff, 1999). To adapt to interviewers’ needs and interview demands, researchers have developed and tested different forms of the CI.

  7.3.3.1 Shortened forms of the CI

  Some researchers have examined shortened forms of the CI, investigating how the full CI compares to versions with various elements removed (Dando, Wilcock, Behnkle, & Milne, 2011; Dando, Wilcock, Milne, & Henry, 2009; Davis, McMahon, & Greenwood, 2005). Studies conducted on shortened forms of the CI have removed elements such as the instruction to recall the event in reverse order or from a different perspective, since research has shown that police officers are likely to abandon these elements because they are difficult to implement. The shortened forms of the CI appear to be as effective as the full CI at increasing the quantity and accuracy of information from interviewees, while also saving time.

  Future research will continue to test this important question: What elements can be removed from the CI without decreasing the quantity or accuracy of recall typically associated with the CI? Answering this question will help law enforcement appropriately allocate time and resources, while also helping scientists learn about how event memory is best enhanced.

  7.3.3.2 Self-administered CI (SAI)

  Many crimes have multiple witnesses. Imagine that instead of a fight between two individuals at the pub, you witness a riot break out in the street after a football match. Dozens are involved. Perhaps a handful of officers arrive on the scene and are faced with the difficult task of sifting through the memory accounts of numerous witnesses. However, they have limited time and resources – where should they start and how should they proceed? It is crucial that they gather as much information as possible, but also that they do so under time constraints because delays could result in witnesses forgetting details, leaving the scene, and/or talking to each other and thus contaminating their memories with others’ recollections. For these reasons, researchers created the SAI (Gabbert, Hope, & Fisher, 2009).

  This variant follows the same basic structure as a standard CI, but requires that interviewees write down their event recall in a booklet instead of conveying their memory accounts aloud to an interviewer. This means that no interviewer needs to be present! Interviewees are asked to proceed through the booklet sequentially and are given instructions to “report everything but avoid guessing”, “reinstate the context”, “d
escribe the perpetrator in detail” and “draw a sketch of the crime scene”. Finally, interviewees are asked to “respond to a series of questions about additional event details that might be important” and to “describe any other witnesses to the crime”.

  Studies have shown that witnesses feel comfortable using the SAI and understand the various instructions. This speaks to the feasibility of using this tool in practice. Notably, the SAI reduces the amount of misinformation that witnesses might otherwise recall if they were interviewed at a later time point (particularly if they talk with other witnesses to the event or listen to the media). The SAI reduces forgetting over time since it can be administered immediately following the event (Gabbert et al., 2009). Importantly, police can still follow up with potential key witnesses after receiving the SAI to gather additional information in the traditional interview context. In fact, the SAI can help police identify who is most beneficial to interview in person. Future research needs to develop variants of the SAI for use with witnesses who may have difficulty reading and writing. Research should also explore the effectiveness of the SAI in eliciting accurate information for different types of crimes.