DON'T MISTAKE A SHORT ANSWER FOR AN EMPTY MEMORY

What a New Study of Forensic Interviews With Preschool-Aged Children Can Teach Us About Asking Better Questions

There is a difficult moment that occurs in many forensic interviews with young children.

The interviewer asks an open-ended question:

"Tell me everything that happened."

The child gives a short answer. Maybe a sentence. Maybe a few words. The interviewer knows there is more information that needs to be explored. Investigators need details. Prosecutors may eventually need specifics. The interviewer may already know information about the allegation that the child has not mentioned.

So the interviewer begins asking more focused questions. That decision makes sense. It may also happen too quickly...

A study by Miriam Johnson and colleagues, first published online in Child Maltreatment in April 2025 and now appearing in Volume 31, Issue 1, provides compelling evidence that preschool-aged children may be capable of providing considerably more forensically relevant information than interviewers sometimes assume.

More importantly, the study suggests that interviewers may frequently move away from open invitations before children have exhausted what they can tell us.

That gives us a simple but important lesson: Don't mistake a short answer for an empty memory!

A STUDY OF REAL FORENSIC INTERVIEWS!!!

Johnson and colleagues examined 256 investigative interviews with preschool-aged children between 2 and 6 years old. These were not laboratory experiments in which children participated in a staged event and were later questioned about it. They were actual investigative interviews involving alleged physical and sexual abuse, and all of the cases proceeded to prosecution.

The researchers focused specifically on "invitations," questions designed to encourage children to retrieve information from their own memory. They distinguished between two forms. A main invitation broadly asks the child to describe what happened, such as asking the child to tell everything that happened. A cued invitation takes something the child has already introduced and asks the child to elaborate on it. For example, if the child mentions going into a bedroom, the interviewer might ask the child to tell more about being in that bedroom.

Both approaches have something important in common. The information originates with the child. That distinction becomes critical in forensic interviewing because the alternative is increasingly focused questioning in which the interviewer begins introducing information, possibilities, people, actions, or choices into the conversation. And the findings from this study were striking.

YOUNG CHILDREN COULD PROVIDE FORENSICALLY RELEVANT INFORMATION

The researchers analyzed 1,065 invitations used during the interviews. Nearly 70% of children's responses to main invitations contained event-specific, forensically relevant information. For cued invitations, that number increased to 83%. Statistically, the researchers did not find a significant difference between the two types of invitations in their ability to elicit forensically relevant information. The children provided details concerning participants, objects, actions, events, locations, and other information relevant to the investigated incident.

Remember the ages of these children... They were 2 to 6 years old!

Children aged 5 and 6 were more likely to elaborate than children aged 2 to 4. That makes developmental sense. Older preschool children generally have greater language abilities and may be better able to organize and communicate their memories. But the study does not support abandoning invitations with younger preschool children. Quite the opposite!

The findings challenge the assumption that preschool-aged children generally require focused questions before they can provide meaningful event-specific information. That distinction matters because our assumptions about children can change the way we question them.

If an interviewer believes a four-year-old cannot provide much information through open-ended questions, the interviewer may quickly begin supplying structure. The child's opportunity to retrieve information freely becomes smaller. Eventually, the interviewer may interpret the child's dependence on focused questions as evidence that focused questions were necessary all along. But we may have created part of the problem ourselves.

THE INTERVIEWERS RARELY USED MAIN INVITATIONS

The researchers identified nearly 40,000 substantive prompts across the 256 interviews. Only 2.7% were invitations. Among the 1,065 invitations identified, only 43 were main invitations. That is 4%!

Even more striking, main invitations appeared in only 32 of the 256 interviews. In other words, approximately 87.5% of these interviews contained no main invitation during the substantive portion of the interview. And when main invitations were used, they often came late. The first main invitation appeared, on average, around the 122nd interviewer-child turn in the substantive phase. The first cued invitation appeared earlier, but still around turn 75 on average. The researchers found that 75% of interviews did not contain their first invitation until after turn 50.

That creates an important question for forensic interviewers: What information might these children have provided earlier if we had simply given them the opportunity?

The researchers found that the likelihood of children's elaboration declined as interviews progressed. For each additional interviewer-child turn, the odds of elaboration decreased by approximately 0.3%.

A fraction of a percent sounds insignificant until we consider how many turns can occur during a lengthy forensic interview. Children have finite cognitive resources. Attention, language production, memory retrieval, emotional regulation, and tolerance for questioning can all become more difficult as an interview progresses. If invitations give us some of our best opportunities to obtain information from the child's own memory, we should use them while the child is best positioned to respond.

THE 93% FINDING SHOULD GET OUR ATTENTION!!!

Perhaps the most practically important finding came from what happened after a child answered a cued invitation. Interviewers rarely followed one invitation with another.

A subsequent cued invitation followed the first cued invitation in only 3.2% of cases. No main invitation was followed by another main or cued invitation. Instead, interviewers frequently moved toward facilitators, directive questions, option-posing questions, and even suggestive questions.

But look at what happened on those relatively rare occasions when interviewers simply gave children another invitation. Children provided additional forensically relevant information 93.4% of the time. That deserves attention!

When interviewers continued asking children to elaborate using information the children themselves had already provided, the children continued producing useful information in more than nine out of ten instances examined. Yet interviewers rarely tried it!

This gives us an important practical lesson: A short response does not necessarily mean the child has nothing else to tell us. Sometimes the appropriate response to a short answer is simply another invitation.

"Tell me more about that."

"Then what happened?"

"You said he went into the room. Tell me everything about what happened when he went into the room."

"You mentioned the car. Tell me more about the car."

These questions do something important. They keep the burden of information production with the child.

THE QUESTION WE ASK NEXT MATTERS

The researchers also examined what happened when interviewers used different types of follow-up questions. The differences were substantial.

When interviewers followed with another cued invitation, children provided additional forensically relevant information 93.4% of the time. When interviewers used directive questions, children elaborated 69.7% of the time. With facilitators, the rate was 68.1%. Then the numbers dropped dramatically.

Following option-posing questions, children provided additional elaboration only 26.1% of the time. Following suggestive questions, it was 28.7%. The type of follow-up question was significantly related to whether children provided additional forensically relevant information.

That means the question we ask next matters!

Consider the difference between two questions...

"Tell me more about what happened in the bedroom."

"Did he touch you when you were in the bedroom?"

The first asks the child to search their memory and provide information. The second introduces an action and asks the child to respond to it. Both questions may eventually have a legitimate place in an interview. There are circumstances where focused questions are necessary to clarify information or address legally important details the child has not provided.

The issue is sequencing. Have we reasonably exhausted what the child can tell us before we begin telling the child what possibilities to consider?

OPEN-ENDED DOES NOT MEAN VAGUE

There is another important lesson here. Sometimes interviewers think of open-ended questions as extremely broad questions such as:

"What happened?"

That is only part of the picture. Cued invitations allow the interviewer to become more focused without introducing new substantive information.

If a child says: "He took me downstairs." The interviewer does not necessarily need to ask: "Did he take you into the bedroom?"

Instead: "You said he took you downstairs. Tell me everything that happened after he took you downstairs."

If the child says: "He gave me something." The interviewer can ask: "You said he gave you something. Tell me more about that."

If the child says: "He told me not to tell." The interviewer can ask: "You said he told you not to tell. Tell me everything he said."

The interviewer is narrowing the inquiry while remaining anchored to information originating from the child. That is an important forensic interviewing skill. It allows us to pursue detail without unnecessarily supplying it.

BETTER QUESTIONS CREATE BETTER INVESTIGATIVE LEADS

There is another reason this research matters beyond the forensic interview itself. The purpose of a forensic interview is not simply to obtain a disclosure. It should help generate information that can be investigated.

Imagine a child independently tells an interviewer: "He took me downstairs after everybody went to bed. The television was still on. He locked the door. My brother was upstairs. He gave me the blue blanket."

Those details give investigators something valuable. They give us questions we can test. Was there a downstairs room? Did the door lock? Was there a television in that location? Was there a blue blanket? Was the brother present? Where was the brother sleeping? Who else was home? Can investigators establish when this could have occurred? Are there photographs of the residence? Are there communications, digital records, location information, security systems, witnesses, or other evidence capable of independently establishing portions of the child's account?

The richer the child's independently generated account becomes, the more opportunities investigators have to corroborate or challenge it. This connects forensic interviewing directly to good investigative practice.

Better questions produce better investigative leads.

Better investigative leads create more opportunities for corroboration.

And better corroboration helps us determine what actually happened.

INVITATIONS ALSO HELP PROTECT THE INTEGRITY OF THE CHILD'S STATEMENT

There is another side to this. Invitations do more than help us obtain additional information. They help us understand where that information came from. This becomes especially important when an allegation is disputed.

Consider the evidentiary difference between a child spontaneously saying: "He locked the door." And an interviewer asking: "Did he lock the door?" followed by the child answering: "Yes."

Those statements may ultimately describe the same alleged fact. But they were produced differently. In the first example, the information originated with the child. In the second, the information originated with the interviewer and was confirmed by the child.

That distinction does not automatically make the second response inaccurate. But it does affect how we should evaluate the source of the information.

The more substantive information children generate themselves, the easier it becomes to identify which details originated from their memory and which details entered the conversation through questioning.

That matters for investigators. It matters for prosecutors. It matters for defense attorneys. And eventually, it may matter to a jury!

FORENSIC INTERVIEWING IS NOT SIMPLY KNOWING A PROTOCOL

Most trained forensic interviewers already know that open-ended questions are preferred. Knowing the principle is relatively easy. Applying it under pressure is harder.

A child gives a two-word response. The interviewer knows important details are missing. A detective is waiting for information. A prosecutor will eventually need specifics. The interviewer may know considerably more about the allegation than the child has mentioned. There is an enormous temptation to help the child get there. That is precisely where interviewer skill matters.

A skilled interviewer develops a repertoire of invitations and learns how to stay with them longer.

"Tell me more."

"Then what happened?"

"What happened next?"

"You said something about the bathroom. Tell me everything about the bathroom."

"You said he came inside. Tell me everything that happened from when he came inside."

The goal is not to endlessly repeat the same question to a child. The goal is to recognize that one brief answer does not necessarily mean free recall has been exhausted.

Johnson and colleagues found a considerable gap between recommended interviewing practices and what occurred in these interviews. Their findings suggest that interviewer perceptions about the limitations of preschool-aged children may contribute to invitations being abandoned prematurely. That is a training problem we can address!

THIS IS WHY FORENSIC INTERVIEWING AND INVESTIGATION CANNOT BE SEPARATED

At the National Crimes Against Children Investigators Association, this is one of the principles built into our Child Sexual Abuse Course, CSAC. We do not want professionals simply memorizing lists of "good questions" and "bad questions." Professionals need to understand why question structure matters.

They need to understand memory retrieval, suggestibility, source monitoring, developmental differences, disclosure dynamics, and the potential effects of repeated or increasingly focused questioning. But they also need to understand what happens after the forensic interview. A well-conducted interview should create investigative opportunities. When a child provides details about people, places, objects, conversations, routines, communications, clothing, vehicles, rooms, gifts, schedules, or events, those details should become investigative leads.

• Can we corroborate them?

• Can we contradict them?

• Can we establish when they occurred?

• Can we identify another witness?

• Can we recover the communication?

• Can we locate the object?

• Can we reconstruct the environment?

• Can we establish opportunity?

• Can we test the timeline?

That is how forensic interviewing becomes part of a larger corroboration-based investigation. And this serves two equally important purposes. It strengthens legitimate cases against offenders. It also reduces the risk of charging an innocent person when the evidence does not support the allegation. Those goals do not compete with one another. They depend on the same thing. Good investigative practice!

DON'T ASSUME THE CHILD HAS NOTHING MORE TO SAY

Johnson and colleagues studied children between 2 and 6 years old in actual criminal investigations. Nearly 70% of responses to main invitations contained forensically relevant information. Cued invitations produced such information in 83% of responses. When interviewers followed one cued invitation with another, children continued providing relevant information 93.4% of the time! Yet main invitations accounted for only 4% of all invitations, approximately 87% of interviews contained no main invitation during the substantive phase, and interviewers rarely followed one invitation with another.

That gap between what children demonstrated they could do and what interviewers gave them the opportunity to do may be the most important finding in the study. Young children have developmental limitations. We should never ignore them. But we should be equally careful about imposing limitations on children that may actually originate from our own expectations.

A four-year-old may give you four words. That does not necessarily mean there are only four words in the child's memory. Ask again. Stay with what the child has already given you. Give the child another opportunity to retrieve the information.

Then listen carefully to what comes next.

Because sometimes the difference between a short statement and a detailed investigative lead is one more well-placed invitation!