Madeleine Bearman, Griffith University
Sonja P. Brubacher, Griffith University
Lydia Timms, Curtin University
Martine Powell, Griffith University

The current study explored the effectiveness of three interview protocols on the number and specificity of details provided by minimally verbal adults about a staged repeated event. Eighty adults (with expressive sentence length of around 5 words, matched on measures of expressive language and intellectual functioning) participated in 3 live events and were pseudorandomly assigned to receive 1 of 3 interviews. The narrative-first protocol exhausted recall with open-ended questions before focused questions were asked, the intermixed protocol paired open-ended questions immediately followed by related focused questions, and the visual cues protocol mimicked the narrative-first protocol but with the use of cue cards. Overall, participants reported more correct information about the last occurrence in the intermixed and visual than narrative-first interview. The narrative-first interview elicited fewer internal intrusions (experienced details attributed to the wrong occurrence) compared with the visual, but not the intermixed, interview. Expressive language and intellectual function were positively associated with the reporting of event related details. Providing information about repeated events was challenging for minimally verbal adults; reporting of generic event details was more frequent than occurrence-specific details, one third of participants answered a question about event frequency incorrectly (by saying they participated once), and the remaining participants provided few details about the other occurrences when directed to do so. Findings were consistent with the broader repeated event literature on children and adult witnesses recalling repeated events. This research provides guidance for investigative interviewers on how best to obtain accurate event-related information from minimally verbal adults about their experiences.

investigative interviewing, minimally verbal adults, repeated events, expressive language difficulty

“Adults with disabilities, including those with communication and/or cognitive impairment, are overrepresented as victims of abuse compared with the general population. Further, they may be at increased risk of repeated victimization, as they tend to be dependent on caregivers who are commonly the perpetrators of abuse.” (p. 239)

“When questioned about their victimization, adults with disability encounter a number of problems in providing evidence to an investigative interviewer unfamiliar with their capabilities. […] Further, investigative interviewers have concerns about the limited detail provided by individuals with communication or cognitive impairment and the reliability of their accounts. These concerns have added to the reduced likelihood of offense allegations proceeding onward from the interview to court or leading to a charge.” (p. 239)

“One particular subgroup of adults with disability that has not been the focus of research on investigative interviewing is minimally verbal adults (those who use sentences of around five words, with or without cognitive impairment). Adults who are minimally verbal are a heterogeneous group and can have a range of complex conditions such as neurodevelopmental disorders (e.g., autism, cerebral palsy), traumatic brain injury, and anatomical/voice difficulties. […] Such individuals often have difficulty combining words to create comprehensible sentences, and their vocabulary is smaller and more basic than their verbally fluent peers, making it more difficult to provide a narrative account. Although a narrative-based protocol is generally advised for all witnesses, it has yet to be determined whether this approach is suitable for minimally verbal adults.” (p. 240)

“Available research suggests that adults who are minimally verbal should be interviewed with open-ended questions, but that they may need some additional support. Approaches to combining open-ended questions with scaffolded support could include pairing open-ended with cued recall questions in an intermixed fashion or using visual aids to remind interviewees of the information being sought.” (p. 241)

“When reporting repeated abuse, witnesses may be required to describe details associated with specific occurrences and avoid confusing details among them. In particular, witnesses are required to give time frames, locations, descriptions of who was there or other unique details specific to an occurrence, as well as the frequency of the offending.” (p. 241)

“Following repeated experience, people develop scripts that characterize the typical features of the event. Scripts are powerful mental representations that help to organize memories and make experiences predictable. It is for this reason that people tend to recall “what usually happens” better than the specific details of any one occurrence. The paradox of repeated event memory is that the very repetition that leads to strong and enduring scripts also hampers the ability to retrieve the correct details associated with individual episodes. Yet the errors that children make when recalling individual episodes are far more likely to be internal intrusions (connecting a true detail that happened at some time to the wrong occurrence) than external intrusions—reports of details that never happened. There is evidence that the same is true for adults.” (p. 241)

“The primary aim of the current study was to determine the interview method that would elicit the most accurate and complete information from minimally verbal adults. Secondary goals involved characterizing the reports of this population with regard to repeated events—in particular, the specificity of details they report and their ability to respond to recommended questions about event frequency. To meet these aims, 80 adults with minimal verbal ability took part in three occurrences of a staged repeated event and then were interviewed using one of three protocols: a narrative-first interview (participants’ narratives were exhausted through open-ended questions before directive questions were asked), an intermixed interview (participants were interviewed using paired nonleading open-ended and cued recall questions in an intermixed fashion), or a visual interview (the same as the narrative-first interview but paired with visual cue cards).” (p. 241)

“Participants comprised 80 minimally verbal adults aged 18 to 74 years (M = 46.93, SD = 14.90; 38 males, 42 females). They had a range of diagnoses, including traumatic brain injury, Down syndrome, cerebral palsy, stroke, intellectual disability, and autism. Participants were included in the study if their guardians (a) nominated them to have expressive language of less than five-word sentences, on average; and (b) expected that they would engage in and recall information about activities. […] Although comprehension was not specifically assessed, all participants were able to respond to a variety of open-ended questions designed to enhance their narrative. Participants were included in the study regardless of whether or not they had a cognitive impairment. They were recruited via the leadership of disability care homes, brain injury services, speech pathology groups, and disability-supported workplaces in one Australian state.” (p. 242)

“Key findings regarding interview type were that some type of nonsuggestive scaffolding provided added benefit to the participants’ accounts compared with an interview without (i.e., narrative-first interview). […] As expected, talking about occurrences of a repeated event was very challenging for this population. They reported more generic than specific details as predicted, but unexpectedly, nearly one third could not effectively answer the “one time or more than one time” question. Of the ones who did answer correctly, 20% were unable to provide details about any other occurrences, and the remainder provided scant information.” (p. 246)

“Participants provided more correct information in the intermixed and visual cues compared with the narrative-first interview. […] Our findings suggest that a free-recall phase may be too broad and that more structure is required to increase narrative information from minimally verbal adults.” (p. 246)

“Contrary to our prediction, participants provided more “don’t know” responses in the narrative-first interview than the visual or intermixed interview. […] Further research is needed to illuminate the reasons for the “don’t know” responding observed in the current study. We propose that the broadness of the questions at the outset of the narrative-first interview, combined with the lack of support provided by visual cues, might have led participants to say, “I don’t know” more often than in the other conditions.” (p. 247)

“A novel aspect of the current research was to describe minimally verbal adults’ reports of repeated events, and it was evident that talking about individual occurrences was challenging. […] Our results suggest that a different approach may be needed to determine how many times an event has occurred when interviewing adults with limited verbal ability and associated cognitive impairments.” (p. 247)

“The current study has reiterated that minimally verbal adults are capable of providing accurate event-related information when interviewed using appropriate questioning techniques, even if the level of detail they can provide is scant. As with other interviewee groups, the onus is on the interviewer to support the interviewee’s ability through effective questioning. The results indicate that all three interview protocols were effective in eliciting event-related detail from the vast majority of individuals who have limited expressive language, and that reporting details specific to occurrences of repeated events is challenging.” (p. 248)

“The present study has demonstrated that, like children and verbally fluent adults, those with limited expressive language report more generic than specific details following repeated experience, and they are prone to internal intrusion errors. These findings are indicative of similar memory organization across the populations and indicate that interviewers should be sensitive to the ways in which they question minimally verbal adults about experiences that may be repeated.” (p. 247)

“To overcome children’s difficulty in reporting occurrence-specific details, some jurisdictions have adopted continuous child sexual abuse statutes, allowing for reduced particularity of a repeated offense. Continuous abuse statues allow for a minimum number of abusive acts to be charged, in relation to a period of time. Currently, in Australia and some jurisdictions of the United States, the complainant must be a young child for a defendant to be charged with continuous abuse. Combined with the prior literature, the findings of the current study provide strong support for the proposition that guidelines for particularization of occurrences be relaxed in multiple vulnerable populations.” (pp. 247–248)

“As participants reported more correct information in the intermixed and visual interview (albeit with increased internal intrusions for the latter) compared with the narrative-first protocol, these more structured interviews would be best used in jurisdictions in which charges of continuous abuse are allowed. For participants in jurisdictions in which such charges are not available, the narrative-first interview may be more appropriate because of lower intrusion errors (at least compared with the visual condition).” (p. 248)

“From a policy standpoint, these findings serve as a reminder that there cannot be a “one size fits all” approach to interviewing. Interviewers and their trainers must be aware of the broader empirical literature that supports various interviewing approaches so that they can tailor best practice protocols according to their needs.” (p. 248)

“The visual interview may be easier to apply than the intermixed interview, as interviewers can utilize open-ended questions by adding visuals, but cues should not be used without training. The intermixed interview is a constant sequence of “Wh” questions paired with open-ended questions. Although it was arguably the best one of the three in the present study, interviewers may find it difficult to alternate between question types without compromising accuracy or slipping into asking solely specific questions unpaired with open-ended ones. […] Interviewers with more developed skills may be able to pair open-ended and specific questions throughout an interview, but for the majority, the visual cues interview would be easier to apply and train.” (p. 248)

“We make the following recommendations for interviewing policy: (a) Charges of continuous abuse should be considered for multiple subcategories of vulnerable witnesses; (b) when continuous abuse charges are allowed, scaffolding (i.e., the visual or intermixed protocol) may help to increase the amount of detail reported; and (c) if particularization is critical, the narrative-first protocol will be the most advantageous because of the lower risk of internal intrusion errors compared with the visual cues interview and greater ease of training compared with the intermixed protocol. The present findings should assist investigative interviewers to recognize both the limitations and capabilities of minimally verbal adults and enhance their ability to obtain essential information about experienced events.” (p. 248)

“Lab studies are critical for understanding memory reports under controlled conditions and provide ground truth. They can be criticized, however, for a lack of generalizability. Although the modified Deakin activities used in the current study bear no resemblance to a criminal offense about which victims might need to be interviewed, the event purposefully includes several details which would have forensic relevance if reported during an interview, such as donning and removal of clothes (leader’s top), body parts to be reported (i.e., during the relaxation), and photographs being shown. Multiple sensory elements (e.g., auditory—music for relaxing; gustatory—snack) are also incorporated.” (p. 248)

“Although differences between the protocols were observed, findings may not generalize to interviews about single experiences or interviews about maltreatment. […] Related to generalizability is the limitation that we did not obtain specific diagnoses for each participant. […] We attempted to overcome this limitation by matching the participants on verbal ability and intellectual function. Conversely, it could be argued that the variability in our participants is a strength of the research because we found differences across interview protocols even with such a heterogeneous sample. We speculate that such heterogeneity would likely be observed among witnesses in field interviews.” (p. 248)

“There were a few limitations associated with our protocol design. Of the three protocols tested, the intermixed condition was the only one in which the main interview differed from that of the practice narrative phase (i.e., only open-ended questions were used in the practice narrative of the intermixed condition). Nevertheless, on the whole, the intermixed condition seemed to have outperformed the other interview conditions. Thus, it is unlikely that this confound impacted the results. Second, interviews in the field may contain leading or suggestive questions, and the impact of such questions on accuracy was beyond the scope of the present study. Finally, neither the modified protocols tested in this study nor the open-ended questioning practices recommended in investigative interviews may be directly applicable to courtroom questioning of witnesses who are minimally verbal. Further work is urgently needed to illuminate the challenges faced by this population during legal proceedings.” (p. 248)

As always, please join the discussion below if you have thoughts or comments to add!