---

Summary of results for the GRACIA Project in-person interview study.

Much research has focused on the vulnerabilities of children who are interviewed in criminal and care and protection investigations. In these interviews, the level of detail and accuracy of information that children provide is critical to understanding what has happened and to inform decision making going forward. Because interviewers were not present when the event or events in question occurred, they may ask questions that children cannot or should not answer. Prior research has established that sometimes children will answer questions that they should not –  because they do not know or remember the answer, they don’t understand the question they’ve been asked, or because the question contains information that is incorrect. In response to this, conversational ‘ground rules’ have been developed, and are explained to children at the beginning of an interview. The rules are:

• to say ‘I don’t know’ when you don’t have the information to answer a question
• to say ‘I don’t understand’ when a question is difficult to comprehend, and
• to correct the interviewer if they make a mistake (‘correct me’)

Currently, interviews with adults do not include ground rule instruction as less attention has been paid to whether adults may also answer questions they shouldn’t. Although adults possess many of the cognitive skills that children are still developing, they might still try to answer questions rather than signal a problem with them. Adults may be motivated to please or comply with an interviewer’s questioning, to appear helpful and competent. They may also feel uncertain about the communicative expectations of the interview, and what types of responses will be viewed as appropriate and acceptable by an interviewer.

To investigate whether adults may also need some support in responding the questions they cannot answer, we looked at their understanding and spontaneous use of conversational ground rules.

Here’s what we did.

One hundred and one students participated in this research as part of Victoria University of Wellington’s School of Psychology’s Introduction to Psychological Research Programme. The students took part in one-on-one interviews with a researcher and completed three tasks.

The Explanation Task required them to help teach ‘Abby the Alien’ how to answer questions. They were asked to explain when each ground rule should and shouldn’t be used – for example, “Tell Abby why people say, ‘I don’t know’ to a question”. This task was designed to assess our participants understanding of the ground rules.

In the Story Task, participants were asked questions about a short story they were read, and the Morning Task involved responding to questions about their own mornings. These tasks included questions which were unanswerable – either because participants didn’t know the answer, couldn’t understand the question, or because the question contained incorrect information. These questions were designed to test how often participants would attempt to answer questions they shouldn’t, and how often they would spontaneously use ground rules.

Here’s what we found.

Across the tasks, we found adults did not express a full understanding of the meaning of ground rules and that they often attempted to answer questions they couldn’t or shouldn’t have. We found ground rule performance was best when participants responded to questions about their own experiences in the Morning Task, compared to questions about someone else’s experiences in the Story Task and questions about the meaning of ground rules in the Explanation Task.

We found adults were best at explaining the ‘don’t know’ rule and showed poorer performance explaining the ‘don’t understand’ and ‘correct me’ rules. In the Story and Morning Tasks, participants were better at correcting the interviewer when she made a mistake, than expressing a lack of knowledge or understanding to answer her questions.

We suspect adults may have performed best in the Morning Task as they were talking about their own experiences, whereas the Story Task described the fictional experiences of another person. In the Explanation Task, participants were not provided with any particular context to consider their responses in (e.g., a criminal investigation), and we observed many responses which explained the ground rules in terms of their function in social exchanges or relationships.

Relatedly, we suggest rule use might have been best for ‘correct me’ as participants had information available to them in the story or based on their own recollection of their morning to recognise that the interviewer had made a mistake. It may have been more challenging for adults to assess and report their lack of knowledge to avoid speculating, and to assess their own comprehension to avoid inferring the meaning of words they did not understand.

These results provide preliminary evidence that there may be some benefit to teaching adults ground rules in interview contexts. We need to investigate further, across different tasks and interview contexts, to build up an evidence base to determine when adults may attempt to answer questions they shouldn’t, and how we can best teach adults to use ground rules appropriately.

Our thanks for your support.

We thank all of those involved in The GRACIA Project for making this research possible, especially the students who agreed to participate in this study.