Dissertations, Theses, and Capstone Projects

Date of Degree

9-2024

Document Type

Dissertation

Degree Name

Ph.D.

Program

Psychology

Advisor

Kelly McWilliams

Committee Members

Deryn Strange

Charles B. Stone

Stacia N. Stolzenberg

Shanna Williams

Subject Categories

Child Psychology | Cognitive Psychology | Developmental Psychology | Law and Psychology | Other Psychology

Keywords

child forensic interviewing; child testimony; event memory; event boundaries

Abstract

Rationale: When a child is a witness or victim of a crime, they are often asked a barrage of questions regarding when the crime took place, how many times it occurred, and how the abuse unfolded (Cross & Whitcomb, 2017; Sullivan et al., 2022). These questions can be problematic because they rely on concepts for which children may lack memory or linguistic ability to answer (Fenson et al., 2000; Friedman, 1993). One fundamental element missing from the copious research (e.g., Brubacher et al., 2012; Stolzenberg et al., 2017a; Wylie et al., 2021) on children’s ability to recount forensically-relevant information is the impact of event boundary formation on the accuracy and reliability of children’s memory reports. We use event boundaries, which mark when an event begins and ends, to make sense of our experiences and predict what will happen next (Zacks et al., 2007). A child's event boundary formation could influence their encoding during an event and subsequent recall during interviewing. Suppose a child does not encode a piece of forensically-relevant information as part of the abusive event. In that case, they cannot recall it when asked about that abusive event later. Additionally, specific questions about forensically-relevant information, such as inter- and intra-numerosity (i.e., how many times something happened across events and within a single event) or mechanics of abuse (i.e., the dynamics of how the abuse unfolded) may cause miscommunication between the child and the questioner if they are not placing boundaries in the same place. For example, a child might interpret the “abuse event” as any abuse that had taken place that day, while the interviewer might be referring only to wrongdoings that occurred in the bedroom at a specific time. Miscommunications regarding important elements of an abusive event such a how many times an action occurred or how the dynamics played out can affect the prosecution’s charging decisions (e.g., Course of Sexual Conduct Against a Child in the First Degree, NYS Penal Law § 130.75 (1) (b), May 2018), the defendant’s right against double jeopardy (Mullane v Central Hanover Bank Trust & Co., 1950), and the child’s credibility (State v. Emmett, 1992). One important mechanism in event boundary formation for adults is analyzing someone else’s intentions (i.e., identifying intentionality, or goal-directed behavior; Zacks, 2004; 2010),which children as young as four are able to do under certain circumstances (Ma & Ganea, 2010). Although event boundaries, intentionality, numerosity, mechanics of abuse, and child interviewing have been studied independently, they are all intertwined when a child is a witness or a victim of a crime. Therefore, this dissertation aims to explore the interconnections of these topics.

Method: I conducted a study to examine how intentionality affected preschool-aged children’s memory for forensically-relevant information (i.e., numerosity and mechanics of abuse) during staged laboratory event using a post-event interview. This study followed a 2 (event: intentionality, control) X 2 (interview: intentionality, control) between-subjects design.

Children in the event intentionality condition were told they were going on a treasure hunt, for which a series of activities were labeled and given goal-directed instructions (i.e., “In order to continue on our treasure hunt, you must complete this Feat of Strength to prove you are strong enough to continue. To complete this Feat of Strength, you must blow up this balloon as much as you can in one minute. Please blow up this balloon as much as you can in one minute”). Following each task, participants received a clue prompting the next task. Children in the event control condition completed the same event, but were simply told they were “going to do some things” and not given labels, explanations, or clues (i.e., “Please blow up this balloon as much as you can in one minute”). Outer clothing (i.e., crown, tutu, and pinny) of both the child and the accompanying research assistant were moved throughout the event to measure clothing placement. Actions were repeated within and across activities to measure inter- and intra-numerosity.

Following the lab event, children completed the NIH Toolbox’s Early Childhood Cognitive Battery and a post-event interview that included questions about numerosity (i.e., “How many times did you stand on one leg?”), their clothing placement (i.e., “Was your tutu on or off?”), their bodily positioning (i.e., “Where was your hand?”), and the order of events (i.e., “Tell me what happened first.”). Various question types were compared for numerosity (i.e., ‘how many’ and ‘once/more than once’), clothing placement (i.e., ‘where,’ ‘on/off,’ and ‘how’), and mechanics of abuse (i.e., ‘inside/outside’ and ‘where’) prompts. Children in the interview intentionality condition were asked questions using the same labels as the event intentionality condition (i.e., “While you competed in the Feat of Strength, where was your pinny?”), while the interview control condition was not (i.e., “Where was your pinny?”).

Results: Given the small sample size and limited power to detect results, all findings must be interpreted with caution. First, I found that children in the intentionality conditions at event and interview had higher accuracy rates and count of events than those in the control conditions, although this difference was not statistically significant. Question type differences replicated previous findings about young children’s struggle with providing specificity themselves (Roberts et al., 2015; Wylie et al., 2021). Second, there was an age effect for children’s invitation response accuracy, but not for direct questions or count of events. Third, children had a particularly low count of events score when they were in the event control and intentionality interview conditions, although this interaction was not statistically significant. Additionally, there was no interactions between conditions for invitation or direct question accuracy. Fourth, there was no statistically significant difference between event conditions for children’s first and last boundaries. Furthermore, while the results did show visual signs of a U-shaped curve, no statistically significant difference between event conditions for children’s responses to the “most important” prompt. While the most common event activity reported was the end goal (i.e., unlocking the treasure chest), children more often reported information about activities outside the bounds of the staged laboratory event, including things they did or were going to do outside the laboratory.

Implications: With my current sample, I found no statistically significant differences in the intentionality conditions, but there are trends that may become clearer with the full sample. Replicating previous findings (Roberts et al., 2015), question type, such as “how many” vs “once/more than once,” impacted children’s accuracy. Moreover, fledgling trends suggest prompts about what happened first, last, and what was most important may be able to be used by questioners to elicit boundary labels from children. With a full sample, this study provides theoretical information about the mechanism behind children’s boundary formation as well as practical legal information about the way boundaries and intentionality impact children’s ability to report forensically-relevant information.

Download

This work is embargoed and will be available for download on Wednesday, August 05, 2026

Share

COinS