“I Didn’t Do That!” Event Valence and Child Age Influence Adults’ Discernment of Preschoolers’ True and False Statements

Abstract

Justice can hinge on adults’ abilities to distinguish accurate from inaccurate child testimony. Yet relatively little is known about factors that affect adults’ abilities to determine the accuracy of children’s eyewitness reports. In this study, adults (N = 108) viewed videoclips of 3- and 5-year-olds answering open-ended and leading questions about positive and negative actually experienced (“true”) events or never experienced (“false”) events that the children either affirmed or denied. Analyses revealed that adults were more accurate at determining the veracity of negative compared with positive incidents, particularly when children said that they had experienced the event. Moreover, adults’ accuracy was at chance for older children’s false denials. Psycholegal implications are discussed.

Keywords

children valence discernment adults accuracy true memory false memory

Children’s participation in the legal system as eyewitnesses has received considerable attention over the last four decades (e.g., Quas & Goodman, 2012; Weisz, Wingrove, & Faith-Slaker, 2007). When children are interviewed in legal cases, they are typically questioned about positive life events (e.g., birthday parties) during the rapport building phase of forensic interviews and about negative life events (e.g., child sexual abuse) during the substantive questioning phase (Lamb, Orbach, Hershkowitz, Esplin, & Horowitz, 2007; Saywitz & Camparo, 2014). Children’s accuracy in response to the latter is of particular legal interest because information acquired at that stage is often essential for criminal justice and child protection investigations. Thus, research on factors that influence children’s eyewitness accuracy about negative life events has burgeoned (see Bruck & Ceci, 1999; Malloy, Johnson, & Goodman, 2013; McWilliams, Bederian-Gardner, Hobbs, Bakanosky, & Goodman, 2012, for reviews). Of significant interest has been preschool children’s (e.g., 3- to 5-year-olds) tendencies to falsely report entire negative events that they never experienced (e.g., Ceci, Loftus, Leichtman, & Bruck, 1994; Schaaf, Alexander, & Goodman, 2008). Yet relatively little is known about factors that influence how adults make decisions about the veracity of younger compared with older preschool children’s reports of true and false events that vary in event valence (negative vs. positive experiences).

As Wigmore (1909) observed over 100 years ago, the legal system can tolerate inaccuracies in eyewitness memory as long as fact finders can ultimately determine the truth. However, research by Block et al. (2012) demonstrated that when adults evaluate preschool children’s reports of past events, adults tend to be fairly accurate when assessing the validity of the children’s true reports (children’s correct affirmations), true denials (children’s correct denials), and false reports (children’s incorrect affirmations), but these same adults are quite deficient in identifying children’s false denials (children’s incorrect denials). From an applied perspective, false denials would occur when children have experienced events (e.g., child abuse) but deny that such events happened. Block et al.’s findings suggest that adults tend to overbelieve such denials.

The current study is an extension of Block et al.’s (2012) research. We attempted to replicate the previous findings as well as examine important factors that may affect adults’ evaluations of children’s statements. Specifically, we focused on two factors (event valence and child age) that could influence the accuracy of adults’ judgments when determining the veracity of children’s statements.

Event Valence

Children are typically called to testify about negative events (Cross & Whitcomb, 2017; Goodman, Quas, Bulkley, & Shapiro, 1999). Yet, to our knowledge, no published research exists on whether event valence influences how accurately adults assess children’s true and false statements. For example, in the Block et al. (2012) study, in which adults faltered in identifying children’s false denials, the researchers did not separately investigate adults’ evaluations of children’s reports about positive and negative experiences.

The existing, limited literature that examines the effects of valence on adults’ decisions about the veracity of eyewitness statements offers two prospects. The first is that adults may have more difficulty accurately judging children’s denials of negative versus positive events. Although surveys reveal that a majority of adults believe that memory is better for dramatic negative events than for more mundane information (e.g., Magnussen et al., 2006), adults likely also consider social factors that can affect adults’ and children’s lack of disclosure of negative compared with positive experiences, especially if disclosures of negative events would be embarrassing or self-incriminating (Bidrose & Goodman, 2000; Leander, Christianson, & Granhag, 2007; Pasupathi, McLean, & Weeks, 2009). For example, in studying adults’ disclosures of life experiences, Pasupathi et al. (2009) found that previously nondisclosed (e.g., omitted or denied) events were significantly more emotionally negative and less emotionally positive than previously disclosed events. Moreover, many adults believe that interviewers should ask children leading questions to elicit disclosure of negative experiences, such as child sexual abuse (Buck, Warren, Bruck, & Kuehnle, 2014; Quas, Thompson, & Clarke-Stewart, 2005). These studies suggest that adults are sensitive to the possibility that children, like adults, may not readily disclose negative self-relevant information. Adults, when considering such social factors affecting children’s disclosures, may find it especially difficult to distinguish between children’s accurate denials (i.e., it really did not happen) and false denials (e.g., the children experienced it, but just don’t want to tell) of negative information. These same social factors are less likely to affect disclosure of positively valenced experiences (Pasupathi et al., 2009). Based on such past research, one could predict that adults would be less accurate at judging denials of negative (compared with positive) events described by children.

Another prospect, however, is that adults may expect different types of details reported for positive versus negative experiences. That is, individuals may rely on heuristics (e.g., general expectations or gist reports of typical routines) when evaluating memories of positive events (e.g., what typically happens at birthday parties) but rely on the presence of more specific or unique details when evaluating memories of negative events (e.g., what specifically happened when punished for misbehavior; Levine & Pizarro, 2004). There is evidence that in evaluating their own memories, adults rely on such heuristics (Kensinger & Schacter, 2006). As a result, adults might use more lenient criteria for evaluating positive events and more stringent, conservative criteria for evaluating negative ones (Levine & Bluck, 2004). An absence of specific detail in children’s false affirmations of negative events, for example, could lead to adults judging children’s false allegations accurately as false reports (e.g., expecting to hear specific negative details, not hearing them, and therefore judging the report as a false allegation). Morever, positive events, if judged more leniently, may be more likely to be overbelieved compared with negative events, resulting in less accuracy in judging memory reports of positive events. This prospect leads to the prediction that adults would be more accurate at judging negative (compared with positive) false affirmations of events as described by children.

We integrated these two prospects by hypothesizing that adults would be more accurate at judging children’s affirmations of negative compared to positive false events but would be less accurate at judging children’s denials of negative false events compared to positive ones.

Children’s Age

Adults expect children to be more accurate with age (Brigham, 1998), assuming that there is no motivation for children to intentionally lie. The increase in perceived accuracy is especially dramatic for children between the ages of 3 and 5 years (Wright, Hanoteau, Parkinson, & Tatham, 2010). Consistent with adults’ intuitions (Brigham, 1998), young preschoolers compared with older children tend to be less accurate on average, to say less, to be more suggestible, and to be more difficult to understand, more variable, and more inconsistent in their reporting of event memories (e.g., Fivush, 1993; Goodman & Reed, 1986; McWilliams et al., 2012). This response pattern may make it difficult for adults to proficiently assess young preschool children’s reports. In contrast, to the extent that adults expect older compared with younger children to recount their past experiences more competently, adults should particularly believe older children’s memory reports, regardless of whether the children are affirming or denying an event’s occurrence and possibly regardless of the older children’s accuracy. However, the greater detail of older children’s reports, especially when they are affirming rather than denying events, may provide information that adults can use, at least to a certain extent, to determine whether the older children are accurate or not. Taken together, we expected that adults would more accurately evaluate older compared with younger children’s affirmations; however, we expected older children’s false denials to be particularly difficult for adults to identify correctly.

The Present Study

Our study examined how accurately adults can detect the veracity of 3- and 5-year-olds’ statements. Participants viewed videoclips previously adapted to address whether adults can distinguish between true and false accounts by children (Block et al., 2012; Schaaf et al., 2008). We presented participants with four possible types of child accounts in which events were either true or false and children either affirmed or denied that the events had happened. Thus, participants were exposed to true affirmations, false affirmations, true denials, and false denials.

For this study, we examined whether event valence and child age influence how well adults judged the accuracy of children’s true versus false memory reports. We focused on adults’ accuracy using a hit minus false alarm rate calculation. Based on the past literature, we advanced the following hypotheses:

Hypothesis 1: Regarding event valence, as explained above, adults would be more accurate at judging false reports and less accurate at judging false denials of negative compared with positive events.

Hypothesis 2: Regarding child age, older children’s denials would be more strongly believed than younger children’s denials due to adults’ expectations of children’s developing abilities to accurately report their past experiences (Neal, Christiansen, Bornstein, & Robicheaux, 2012). However, given deficits in how well 3-year-olds answer memory questions, we expected adults to have difficulty accurately assessing young preschool children’s affirmations compared to older children’s affirmations.

Method

Participants

Participants were 108 undergraduate students (62 female), aged 18 to 58 years (M = 20.53, SD = 4.18) from a large research university located in the western United States. Participants’ ethnicities were diverse, (37% Caucasian non-Hispanic, 41% Asian/Asian American, 18% Hispanic/Latino, 2% African American, and 2% “other”) consistent with the university’s student population.

Materials

Demographics form

A short questionnaire was constructed to collect typical demographic information (e.g., participant gender).

Videoclips

Videoclips from a previous study (Schaaf et al., 2008) served as stimuli. In these videoclips, children were interviewed about true and false events that were positive or negative in nature. For the previous study, parents of the children had been asked to identify true events that their children experienced and false events that the children had not experienced. False events came from the pool of true events that other children had experienced, as reported by the parents. That is, a false event for one child was often a true event for another child.

In the Schaaf et al. (2008) study, the child interviews contained four open-ended questions and two yes–no questions about each event (e.g., “Tell me about the time you poopied in your pants.” “Was your mom there when you poopied in your pants?” [a true event]), standardized in terms of question type and tailored individually for the experiences (based on parent report) of the child interviewed. Children were interviewed about two types of positive events (i.e., times when children received gifts or went on trips) and two types of negative ones (i.e., times when children experienced embarrassment or punishments).

For the Block et al.’s (2012) study, 24 of the Schaaf et al.’s (2008) videoclips, plus one practice videoclip (25 different children), were randomly selected, with the restriction that the children’s responses evenly fell into four narrative types, as described above: true affirmations, true denials, false affirmations, and false denials. These same 25 videoclips were used in the current study. As in Block et al.’s research, there were no significant differences in number of words provided by the interviewers or the children across the four narrative types, and there were no significant differences in event plausibility ratings across these four categories (see Block et al., 2012; Schaaf et al., 2008, for details).

Children were divided evenly by age for each of the four narrative types (e.g., three 3-year-olds and three 5-year-olds who made false denials). Child gender and event valence were approximately equally represented for each of the four categories. The interviews consisted of five positive and seven negative true events and seven positive and five negative false events. The child’s age and gender appeared (in written form) on the videoclips before each child was shown answering questions.

Ethnicities of children featured in the videoclips were 17 Caucasian non-Hispanic, one Hispanic/Latino, two Asian, and five other/mixed. The practice videoclip, which appeared first, featured a 3-year-old Caucasian boy who made a true affirmation about a negative event. The order of the remaining 24 videoclips was randomized, with the exception that no more than three child affirmations or denials and no more than three of the same event category (e.g., gift, trip, trip) were presented in a row. Order effects were not observed in Block et al.’s (2012) study, which used the same videoclips as we did. In the present study, all participants saw the same randomized stimulus order.

Children’s Statement Evaluation Questionnaire

This questionnaire, for completion by adult participants, included questions about the practice interview and about each child interview. Regarding each child, the first question concerned whether or not the event in question happened (yes or no). The other questions per child duplicated those described by Block et al. (2012); these additional questions assessed perceptions about each child (e.g., accuracy, suggestibility). However, the findings largely duplicated those reported by Block et al. As our concern in the present article was with hit minus false alarm scores, results for the additional questions are not reported.

Procedure

The study was approved by the university’s Institutional Review Board and took place in a large lecture hall on the university campus. Participants came in groups of 20 to 30 per session. Videoclips were displayed on a 16-foot screen at the front of the room. All participants were told (accurately) that the interviewers did not know which events the children had experienced and which events the children had not experienced.

Participants were instructed to take the task seriously, just as they would for a legal case. They then viewed the practice clip and answered the practice questions on the Children’s Statement Evaluation Questionnaire. Next, they viewed one child interview at a time and answered the relevant questions about that interview before moving on to the next child interview. Participants subsequently completed a demographics form and were debriefed and thanked for their contributions. No feedback on the accuracy of the children’s responses, including of the practice clip, was provided. The entire task lasted approximately 1 hr.

Results

Preliminary analyses indicated no significant main effects or interactions involving participant gender. However, given consistent gender differences in relevant past research (Bottoms et al., 2014), we covaried gender in our main analyses. Due to the limited number of videoclips per cell of our full factorial design, we statistically examined effects of event valence and child age in separate analyses of covariance (ANCOVAs). Specifically, participant hit minus false alarm rates in evaluating whether children experienced the events were examined via 2 (event valence: positive vs. negative) × 2 (report index: affirmations vs. denials) repeated-measures ANCOVAs, with valence and report index as within-subject factors and participant gender as the covariate. In separate ANCOVAs, the variable of event valence was replaced with child age. Significant interactions were dissected into simple effects. All significant effects are reported.

Proportion Accuracy

Although our main interest was in hit minus false alarm scores, it is instructive first to informally consider the pattern of means for adults’ proportion correct responses (Table 1). Inferential statistics on these scores are not presented here. However, in regard to the four event categories (e.g., false denials), Block et al.’s (2012) pattern of findings was replicated (e.g., adults were especially accurate in identifying true denials and especially inaccurate in identifying false denials). New here is that adults appeared more accurate in judging the veracity of children’s responses about negative compared with positive events that were false. Adults were also apparently more accurate when 5- compared with 3-year-olds made true affirmations. However, the age trend was reversed when children made false denials: Adults were less accurate identifying 5- compared with 3-year-olds’ false denials.

Table 1.

Mean Proportion Accuracy and Standard Deviations by Event Valence and Child Age.

	Event Type
	True Affirmations	False Affirmations	True Denials	False Denials
Event valence
Positive	.62 (.25)	.63 (.27)	.97 (.15)	.06 (.20)
Negative	.72 (.26)	.83 (.28)	.96 (.10)	.21 (.21)
Child age
3-year-olds	.52 (.28)	.69 (.29)	.96 (.16)	.26 (.26)
5-year-olds	.82 (.20)	.70 (.28)	.96 (.12)	.05 (.17)

Hit Minus False Alarm Analyses

The limited number of stimuli participants observed per category unfortunately prevented the use of d′ and response bias scores; however, hit minus false alarm scores were calculated incorporating premises of signal detection theory (Macmillan & Creelman, 2005; Wixted, 2007). To obtain the hit minus false alarm scores for adults’ judgments of children’s affirmations, we first calculated individual hit and false alarm rates for each adult participant regarding the judgment of whether or not the event affirmed by the child was experienced by the child. For affirmations, a hit was defined as when the participant correctly identified that the event being affirmed by the child actually occurred. A false alarm was defined as when the adult participant indicated that the event had been experienced when the child falsely affirmed the event. Next, for each participant, we subtracted the false alarm rate from the hit rate to obtain a hit minus false alarm score. A comparable procedure was used to calculate hit minus false alarm scores for children’s denials: Specifically, hits for denials were defined as adult participants correctly judging that an event had not happened when the child denied that the event had occurred, and false alarms were defined as participants falsely judging that an event had not happened when the child denied that a truly experienced event had occurred. A score less than 0 indicates a lack of accurately discriminating the children’s true versus false answers. For example, a hit minus false alarm score of –.55 would mean that the participant had more false alarms compared with hits, performing poorly in assessing the accuracy of the children’s reports. Note that creation of these scores incorporates comparison of true versus false statements, eliminating the true versus false report variable in the analyses.

With hit minus false alarm scores as the dependent measure (Table 2), ANCOVA revealed significant main effects for event valence, F(1, 106) = 50.83, p < .001, $η_{p}^{2}$ = .32, and report index, F(1, 106) = 76.49, p < .001, $η_{p}^{2}$ = .42, and a trend for their predicted interaction, F(1, 106) = 3.82, p = .05, $η_{p}^{2}$ = .04. Given we hypothesized this interaction, we dissected it into simple effects. These indicated that participants’ scores for negative events that children affirmed (“Yes, I threw a rock through a window”) were significantly higher than their scores for positive events that children affirmed (e.g., “Yes, I went to Disneyland”), F(1, 106) = 32.41, p < .001, $η_{p}^{2}$ = .23. Similarly negative denials were judged with greater accuracy compared with positive denials, F(1, 106) = 22.50, p < .001, $η_{p}^{2}$ = .18. Moreover, participants’ scores for positive events that children denied were not statistically different from chance, t(107) = −0.18, ns; all other scores were significantly above the chance level of 0, t(107) ≥ 7.52, p < .001. Thus, participants had particular difficulty assessing children’s accuracy for positive events that the children denied experiencing.

Table 2.

Mean H – FA Scores and Standard Deviations by Event Valence and Child Age.

	H – FA Affirmations	H – FA Denials	M (SD)
Event valence
Positive	.24 (.27)	−.003 (.18)	.12 (.12)
Negative	.54 (.35)	.17 (.24)	.36 (.15)
M (SD)	.39 (.15)	.08 (.12)
Child age
3-year-olds	.21 (.36)	.23 (.27)	.22 (.16)
5-year-olds	.51 (.29)	.00 (.18)	.26 (.13)
M (SD)	.36 (.16)	.11 (.13)

Note. H – FA = hit minus false alarm.

We then replaced valence with child age in the ANCOVA. A significant Child Age × Report Index interaction emerged, F(1, 106) = 78.12, p < .001, $η_{p}^{2}$ = .42. Simple effects indicated that participants were better at assessing the accuracy of 5- than 3-year-olds’ affirmations, F(1,106) = 33.22, p < .001, $η_{p}^{2}$ = .24. Yet the reverse was observed for denials: Participants were better at detecting the accuracy of 3- compared with 5-year-olds’ denials, F(1,106) = 38.25, p < .001, $η_{p}^{2}$ = .27. Scores for 3-year-olds’ affirmations, 3-year-olds’ denials, and 5-year-olds’ affirmations were statistically different from the chance value of 0, t(107) ≥ 6.00, p’s < .001. However, participants scores for 5-year-olds’ denials did not differ significantly from chance, t(107) = 0.00, ns.

Discussion

We investigated factors that influence how adults evaluate preschool children’s true and false affirmations and denials. We expanded on Block et al.’s (2012) findings by considering the effects of event valence and child age on adults’ abilities to make accurate decisions about true and false events described by young children. For proportion accuracy, the pattern of means for the four narrative types (true affirmations, false affirmations, true denials, false denials) replicated that in the Block et al. study, although adults were most accurate when they evaluated true denials, adults also were relatively accurate in judging when children made true affirmations (aka, “true reports”) and false affirmations (aka, “false reports”). Regarding errors preschoolers may make, the latter finding is particularly encouraging as it implies, in contrast to findings by Leichtman and Ceci (1995), that adults might fairly readily detect false reports of events. However, children who falsely denied experiencing an event when it had actually occurred (as confirmed by parents) were judged incorrectly. To the extent that these findings generalize to legal contexts, children who falsely deny experiences, intentionally or unintentionally, may fail to disclose events that are of interest to authorities, limiting how well the legal system can deliver justice for potentially significant harms.

Event Valence

Adults generally believe that memory is better for dramatic negative experiences than mundane ones (Magnussen et al., 2006). However, adults also realize that social factors, such as embarrassment and self-incrimination, can affect children’s disclosures (Buck et al., 2014; Quas et al., 2005). Given that the children in our study were asked about such negative experiences as having their pants pulled down in front of others (an embarrassing event) or getting in trouble for throwing a rock through a window (a self-incriminating event), it was of interest to examine whether adults perceived children’s affirmations or denials of such events as more or less accurate than when the children were interviewed about positive events (e.g., getting a stuffed animal hippo for their birthday).

Analyses indicated that when children described negative (compared with positive) events, adults were more accurate in judging whether these events occurred. In fact, this was true when children denied events, although scores for both report index categories (affirmations and denials) were significantly higher for negative than positive event experiences, as indicated in the simple effects analyses. It is reassuring that adults are better at judging negative events given that, when children are involved in the legal system, their affirmation of negative events is typically at issue (Goodman et al., 1999). Moreover, adults were above chance for all event valence analyses categories except for positive events that children denied experiencing. Again, however, the hit minus false alarm scores were lower than what would be ideal: The average hit minus false alarm scores for negative affirmations and negative denials were .54 and .17, respectively. These values were significantly different from chance levels; however, they are still relatively low, especially for the negative denials. As applied to children’s involvement in the legal system, these findings suggest that if children make false denials about negative events (e.g., falsely deny being sexually abused), adults, in some if not many instances, may believe these denials and potentially inadvertently prolong children’s victimization and delay access to legal and mental health services.

Past research on adults’ assessments of their own memories indicates that negative events are evaluated more stringently and with more conservative tendencies than are positive events (e.g., Levine & Bluck, 2004). It was therefore possible that adults would have similar biases about children’s reports. When children indicated that positive events occurred, participants may have anticipated hearing heuristically appropriate details (e.g., gist-level information that matches general expectations about the events) that young children (especially 3-year-olds) do not necessarily report, leading adults to reject that the event occurred, regardless of whether the affirmation was true or false. In addition, expecting specific or unique details for negative affirmations (but not negative denials), and not hearing said details for such affirmations, may have improved participants’ abilities to accurately judge negative false reports compared with positive false reports (Levine & Bluck, 2004); this would ultimately increase participants’ hits minus false alarm scores for negative experiences, as the adults would have made fewer false alarms for negative compared with positive events. We unfortunately did not assess participants’ reasoning behind why they believed some children’s affirmations but not others, yet the results imply that event valence accounted for part of this decision. Further research is needed to examine such possibilities.

The majority of children gave short responses to the questions, but the content of the answers differed between affirmations and denials. In particular, children frequently failed to provide contextual or perceptual details about the events; rather, when they denied an experience, they often (but not always) simply denied that the events occurred (regardless whether the denial was true or false), for example, by saying “Nope, that didn’t happen,” or “I never went there,” whereas an affirmation might have been (regarding a trip to Disneyland) “Just bought a hat.” Children who talked more were often off topic. The adults likely could not easily evaluate or monitor details that were contextually relevant to the positive or negative events that were denied; thus, denials may have been particularly difficult to evaluate accurately. Instead, participants may have relied on the children’s nonverbal cues and the plausibility of the events being described (Qin, Ogle, & Goodman, 2008; Talwar, Lee, Bala, & Lindsay, 2006).

Children’s Age

We examined whether adults were more accurate in evaluating children’s statements depending on the children’s age. Adults expect children to provide more accurate memory reports as children mature (e.g., Brigham, 1998; Neal et al., 2012), especially for children in the age ranges represented in our study (Wright et al., 2010; but see Talwar et al., 2006). Past research also shows that adults overbelieve children’s false denials (Block et al., 2012). Taken together, these findings led us to predict that adults would believe denials made by older children more than ones made by younger children. We also expected that adults would be more accurate in assessing older compared with younger children’s affirmations.

In prior studies, adults had difficulty detecting adults’ and children’s false statements (Laimon & Poole, 2008; Orcutt, Goodman, Tobey, Batterman-Faunce, & Thomas, 2001; Westcott, Davies, & Clifford, 1991). These studies typically compared children of older ages or across a wider age range than in the present study (but see Leichtman & Ceci, 1995). We found that adults had the greatest difficulty identifying 5- compared with 3-year-olds’ denials, not their false affirmations. As children age, their abilities to deceive improve (e.g., Talwar & Lee, 2008). Nevertheless, adults perceive children to be more honest as age increases (e.g., Wright et al., 2010). In any case, adults’ overbelief in older children’s denials (particularly false denials) raises concerns as to whether adults are able to recognize false denials made by older children in legal contexts. In line with the past literature, adults’ bias to overbelieve older compared with younger children may have contributed to the adults being less able to identify the 5-year-olds’ falsehoods. Our results cannot be attributed to an overrepresentation of 5-year-olds in one category of children’s statements: We had counterbalanced our videoclips so that participants viewed an equal number of 5-year-olds making affirmations and denials.

Future investigations should examine a wider age range of children. Adults may more readily perceive differences between the reports of preschoolers and older elementary age children (e.g., 7- and 9-year-olds), which is important since the eyewitness ability of children across a wide age range is frequently questioned (e.g., Ceci & Bruck, 1994; Poole & Lindsay, 2001). We did not examine possible age differences in the characteristics of the children’s reports (e.g., false denials; but see Schaaf et al., 2008); as such differences might influence adults’ evaluations, this possibility would be a valuable direction for future research.

Limitations

Although our study expands scientific knowledge regarding how adults evaluate children’s statements, several limitations should be noted. First, we could not examine child age and event valence together in a single analysis. Including more than 24 videoclips was not feasible given our reliance on stimuli from a previously completed study and the potential fatigue involved for adult participants in the current study. The limited number of videotapes per category prevented the use of standard signal detection theory measures (i.e., d′ and response bias): The mathematical correction for mean scores of 1.00 and 0.00 has a large effect when there are relatively few videotapes per category, which would have transformed the data too severely to provide meaningful interpretation. Second, although events falling into our four event categories (e.g., false affirmations, true denials) were judged to be plausible, the negative events that we explored in the present study may be less plausible than the positive events (e.g., the child having her or his pants pulled down vs. taking a trip to Disneyland). Although plausibility differences may be confounded with event valence in real-world contexts, future research should attempt to differentiate between event valence and plausibility. Future studies should also attempt to counterbalance events across categories. Third, parents and researchers rather than children decided the valence designation of the events in Schaaf et al.’s (2008) research, which carried over to the present project. Moreover, the negative events studied fall short in ecological validity in relation to many events of concern in the criminal justice and child protection systems (e.g., child sexual abuse). That said, the findings may still be of relevance to certain cases in legal contexts (including child custody disputes and personal injury actions), wherein children’s descriptions of true and false events, especially events considered negative by adults, may also be questioned. Fourth, young children may be perceived differently than older children depending on the type of act (Neal et al., 2012). Future research should attempt to overcome such limitations.

Practical Applications

Generalizing our findings to a legal context, children who purposefully or inadvertently falsely deny experiences may fail to reveal events that are of interest to legal authorities. These events then may never come to the attention of the legal system or the cases may be dropped. Consistent with Block et al.’s (2012) results, the current findings highlight adults’ difficulty in accurately detecting children’s false denials, particularly as children age. In addition, evaluation of children’s statements about negative events, both for affirmations and denials, were above chance levels although arguably still lower than what one might consider ideal in application. Research is needed on ways to sharpen detection of false denials while not adversely affecting detection of false reports.

Conclusion

In several past studies examining how accurately adults’ evaluate the veracity of children’s narratives, mean accuracy scores were used to assess adults’ performance (e.g., Strömwall, Granhag, & Landström, 2007; Talwar et al., 2006; Warren, Dodd, Raynor, & Peterson, 2012). Using hit minus false alarm scores, we found that reports about negative (vs. positive) events were judged with greater accuracy. In addition, adults had difficulty identifying the veracity of children’s denials, particularly for 5- compared with 3-year-olds. These findings hold relevance for the criminal justice system in two ways: (a) children are typically questioned about negative events that they either affirm or deny and then adults use such information to decide whether to pursue investigations and (b) children’s accuracy, which is frequently called into question, is likely to be particularly scrutinized in the preschool years.

Footnotes

Acknowledgements

The authors thank Dr. Jennifer Schaaf and Dr. Kristen W. Alexander for their prior contributions, and Dr. Ronald Fisher and Dr. John Wixted for their feedback and consultation.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors acknowledge and thank the Amini Foundation for the Study of Affects and the University of California, Davis, School of Law for their support of this research.

Author Biographies

Jonni L. Johnson, MA, is a doctoral student at the University of California, Davis. She received her bachelor’s degree in psychology from the University of Pittsburgh, with a minor in administrative justice. Before attending graduate school, she worked for several years in a laboratory at the University of Pittsburgh that specializes in children with autism. In her doctoral studies, she is focusing on psychology and law, with specific interests in memory development, perceptions of child witnesses, and forensic interviewing (including eyewitness memory in children and adolescents with autism spectrum disorder). She has published empirical and review articles on such topics as children’s advocacy centers, children’s memory and suggestibility, and forensic interviewing. Ms. Johnson is editor of a book on eyewitness memory in individuals with autism spectrum disorder. She has a strong background in quantitative psychology.

Sue D. Hobbs, PhD, is assistant professor of child development in the College of Education at California State University, Sacramento. She received her bachelor’s degree in psychology from California State University, East Bay. She then completed her MA and PhD in developmental psychology at the University of California, Davis. She specializes in research on child victims in the criminal justice and child protection systems, child forensic interviewing, and children in foster care. She has published systematic reviews and meta-analyses of child witness research, as well as publishing numerous other empirical studies and review chapters. She is currently on the Executive Committee of Division 7 (Developmental Psychology) of the American Psychological Association.

Yoojin Chae, PhD, is assistant professor in human development and family studies at Texas Tech University. Her research focuses on children’s abilities as witnesses, including children’s eyewitness memory and suggestibility, trauma and memory, attachment and memory, language and memory, forensic interview techniques, and adults’ evaluations of children’s autobiographical memory. She has published important papers on event memory and suggestibility in maltreated children in relation to trauma-related psychopathology (e.g., dissociation, posttraumatic stress disorder [PTSD], depression) and a series of studies on the link between attachment and memory for distressing events. She is currently studying the effects of interviewing techniques and narrative skill on low-socioeconomic status children’s eyewitness memory. She received her doctorate in human development from Cornell University.

Gail S. Goodman is distinguished professor of psychology at the University of California, Davis. Her research concerns such topics as children’s and adults’ eyewitness memory, child maltreatment, forensic interviewing, victimization, legal involvement, and trauma and memory. In the past, she served as president of several divisions of the American Psychological Association (e.g., Developmental Psychology, American Psychology–Law Society, Child and Family Policy and Practice). She is internationally known for her research on victims and witnesses, which has been cited multiple times by the U.S. Supreme Court. She has published widely and received numerous national awards and grants for her research and writings. She received her PhD in developmental psychology from University of California, Los Angeles, and went on to complete postdoctoral research at the University of Denver and the Université René Descartes in Paris, France. She also holds an honorary doctorate in psychology from the University of Oslo, Norway.

Donna Shestowsky is a tenured professor at the School of Law and an affiliated faculty member of the Department of Psychology at the University of California, Davis. She was awarded bachelor’s and master’s degrees from Yale University and holds a PhD in psychology and JD from Stanford University. During the 2003-2004 academic year, she was jointly appointed to the faculty at Northwestern University School of Law and the Kellogg School of Management. She currently teaches criminal law, negotiation strategy, alternative dispute resolution, and a seminar in legal psychology at the University of California, Davis. Her main research objective is to examine basic assumptions underlying the structure of the legal system and to explore ways in which the legal system might be improved using the methodological and analytic tools of psychological theory and research. She was principal investigator of a multiyear research project, funded by the National Science Foundation and the American Bar Association, to examine how litigants evaluate legal procedures. Her research has been published in top journals in both psychology and law, including the Stanford Law Review, Law and Human Behavior, and the Journal of Personality and Social Psychology.

Stephanie D. Block, PhD, is assistant professor of psychology at the University of Massachusetts Lowell. Her research broadly focuses on children in the legal system, the effect of trauma on children’s well-being and on their memory for emotional events, and the prevention of child maltreatment. Guided by a social-ecological perspective and interdisciplinary training, she conducts research that generates knowledge and informs public policy relevant to children in the child welfare and legal systems. She has published her research in numerous journals, including Law and Human Behavior, Development and Psychopathology, and Behavioral Sciences & the Law. She joined the faculty of the University of Massachusetts Lowell, in the fall of 2011 and teaches courses in psychology and law and in child and adolescent development. She received her PhD in developmental psychology at the University of California, Davis, and completed postdoctoral studies at the University of North Carolina, Chapel Hill.

References

Bidrose

Goodman

G. S.

(2000). Testimony and evidence: A scientific case study of memory for child sexual abuse. Applied Cognitive Psychology, 14, 197-214.

Block

S. D.

Shestowsky

Segovia

D. A.

Goodman

G. S.

Schaaf

J. M.

Alexander

K. W.

(2012). “That never happened”: Adults’ discernment of children’s true and false memory reports. Law and Human Behavior, 36, 365-374. doi:10.1037/h0093920

Bottoms

B. L.

Peter-Hagene

L. C.

Stevenson

M. C.

Wiley

T. R. A.

Schneider Mitchell

Goodman

G. S.

(2014). Explaining gender differences in jurors’ reactions to child sexual assault cases. Behavioral Sciences & the Law, 32, 789-812. doi:10.1002/bsl.2147

Brigham

J. C.

(1998). Adults’ evaluations of characteristics of children’s memory. Journal of Applied Developmental Psychology, 19, 15-39. doi:10.1016/S0193-3973(99)80026-4

Bruck

Ceci

S. J.

(1999). The suggestibility of children’s memory. Annual Review of Psychology, 50, 419-439.

Buck

J. A.

Warren

A. R.

Bruck

Kuehnle

(2014). How common is “common knowledge” about child witnesses among legal professionals? Comparing interviewers, public defenders, and forensic psychologists with laypeople. Behavioral Sciences & the Law, 32, 867-883.

Ceci

S. J.

Bruck

(1994). How reliable are children’s statements? . . . It depends. Family Relations, 43, 255-257.

Ceci

S. J.

Loftus

E. F.

Leichtman

M. D.

Bruck

(1994). The possible role of source misattributions in the creation of false beliefs among preschoolers. International Journal of Clinical and Experimental Hypnosis, 42, 304-320.

Cross

T. P.

Whitcomb

(2017). The practice of prosecuting child maltreatment: Results of an online survey of prosecutors. Child Abuse & Neglect, 69, 20-28. doi:10.1016/j.chiabu.2017.04/007

10.

Fivush

(1993). Developmental perspectives on autobiographical recall. In Goodman

G. S.

Bottoms

B. L.

(Eds.), Child victims, child witnesses: Understanding and improving testimony (pp. 1-24). New York, NY: Guilford Press.

11.

Goodman

G. S.

Quas

Bulkley

Shapiro

(1999). Innovations for child witnesses: A national survey. Psychology, Public Policy, and Law, 5, 255-281.

12.

Goodman

G. S.

Reed

R. S.

(1986). Age differences in eyewitness testimony. Law and Human Behavior, 10, 317-332. doi:10.1007/BF01047344

13.

Kensinger

E. A.

Schacter

D. L.

(2006). When the Red Sox shocked the Yankees: Comparing negative and positive memories. Psychonomic Bulletin & Review, 13, 757-763.

14.

Laimon

R. L.

Poole

D. A.

(2008). Adults usually believe young children: The influence of eliciting questions and suggestibility presentations on perceptions of children’s disclosures. Law and Human Behavior, 32, 489-501.

15.

Lamb

M. E.

Orbach

Hershkowitz

Esplin

P. W.

Horowitz

(2007). A structured forensic interview protocol improves the quality and informativeness of investigative interviews with children: A review of research using the NICHD investigative interview protocol. Child Abuse & Neglect, 31, 1201-1231.

16.

Leander

Christianson

S.-A.

Granhag

P. A.

(2007). A sexual abuse case study: Children’s memories and reports. Psychiatry, Psychology, and Law, 14, 120-129.

17.

Leichtman

Ceci

S. J.

(1995). The effects of stereotypes on suggestions on preschoolers’ reports. Developmental Psychology, 31, 568-578. doi:10.1037//0012-1649.31.5.758

18.

Levine

L. J.

Bluck

(2004). Painting with broad strokes: Happiness and the malleability of event memory. Cognition and Emotion, 18, 559-574. doi:10.1080/02699930341000446

19.

Levine

L. J.

Pizarro

D. A.

(2004). Emotion and memory research: A grumpy overview. Social Cognition, 22, 530-554. doi:10.1521/soco.22.5.530.50767

20.

Macmillan

N. A.

Creelman

C. D.

(2005). Detection theory: A user’s guide (2nd ed.). Mahwah, NJ: Lawrence Erlbaum.

21.

Magnussen

Andersson

Cornoldi

De Beni

Endestad

Goodman

G. S.

. . . Zimmer

(2006). What people believe about memory. Memory, 14, 595-613. doi:10.1080/09658210600646716

22.

Malloy

L. C.

Johnson

J. L.

Goodman

G. S.

(2013). Children’s memory and event reports: The current state of knowledge and best practice. Journal of Forensic Social Work, 3, 106-132.

23.

McWilliams

Bederian-Gardner

Hobbs

S. D.

Bakanosky

Goodman

G. S.

(2012). Children’s eyewitness memory and suggestibility: Revisiting Ceci and Bruck’s (1993) review. In Slater

A. M.

Quinn

P. C.

(Eds.), Developmental psychology: Revisiting the classic studies (pp. 101-117). Thousand Oaks, CA: SAGE.

24.

Neal

T. M. S.

Christiansen

Bornstein

B. H.

Robicheaux

T. R.

(2012). The effects of mock jurors’ beliefs about eyewitness performance on trial judgments. Psychology, Crime & Law, 18, 49-64. doi:10.1080/1068316X.2011.587815

25.

Orcutt

Goodman

G. S.

Tobey

Batterman-Faunce

J. M.

Thomas

(2001). Detecting deception in children’s testimony: Factfinders’ abilities to reach the truth in open court and closed-circuit trials. Law and Human Behavior, 25, 339-372.

26.

Pasupathi

McLean

K. C.

Weeks

(2009). To tell or not to tell: Disclosure and the narrative self. Journal of Personality, 77, 89-124. doi:10.1111/j.1467-6494.2008.00539.x

27.

Poole

D. A.

Lindsay

D. S.

(2001). Children’s eyewitness reports after exposure to misinformation from parents. Journal of Experimental Psychology: Applied, 7, 27-50.

28.

Qin

Ogle

C. M.

Goodman

G. S.

(2008). Adults’ memories of childhood: True and false reports. Journal of Experimental Psychology: Applied, 14, 373-391. doi:10.1037/a0014309

29.

Quas

J. A.

Goodman

G. S.

(2012). Consequences of criminal court involvement for child victims. Psychology, Public Policy, and Law, 18, 392-414. doi:10.1037/a0026146

30.

Quas

J. A.

Thompson

W. C.

Clarke-Stewart

(2005). Do jurors “know” what isn’t so about child witnesses? Law and Human Behavior, 29, 425-456. doi:10.1007/s10979-005-5523-8

31.

Saywitz

K. J.

Camparo

(2014). Evidence-based child forensic interviewing: The developmental narrative elaboration interview. New York, NY: Oxford University Press.

32.

Schaaf

J. M.

Alexander

K. W.

Goodman

G. S.

(2008). Children’s false memory and true disclosure in the face of repeated questions. Journal of Experimental Child Psychology, 100, 157-185. doi:10.1016/j.jecp.2007.09.002

33.

Strömwall

L. A.

Granhag

Landström

(2007). Children’s prepared and unprepared lies: Can adults see through their strategies? Applied Cognitive Psychology, 21, 457-471.

34.

Talwar

Lee

(2008). Social and cognitive correlates of children’s lying behavior. Child Development, 79, 866-881. doi:10.1111/j.1467-8624.2008.01164.x

35.

Talwar

Lee

Bala

Lindsay

R. C. L.

(2006). Adults’ judgments of children’s coached reports. Law and Human Behavior, 30, 561-570. doi:10.1007/s10979-006-9038-8

36.

Warren

K. L.

Dodd

Raynor

Peterson

(2012). Detecting children’s lies: Comparing true accounts about highly stressful injuries with unprepared, prepared, and coached lies. Behavioral Sciences & the Law, 30, 329-341.

37.

Weisz

Wingrove

Faith-Slaker

(2007). Children and procedural justice. Court Review: The Journal of the American Judges Association, 44, 36-44.

38.

Westcott

H. L.

Davies

G. M.

Clifford

B. R.

(1991). Adults’ perceptions of children’s videotaped truthful and deceptive statements. Children & Society, 5(2), 123-135. doi: 10.1111/j.1099-0860.1991.tb00378.x

39.

Wigmore

J. H.

(1909). Professor Münsterberg and the psychology of testimony. Illinois Law Review, 3, 399-445.

40.

Wixted

J. T.

(2007). Signal detection theory and the neuroscience of recognition memory. In Nairne

(Ed.), The foundations of remembering: Essays in honor of Henry L. Roediger III (pp. 67-82). New York, NY: Psychology Press.

41.

Wright

D. B.

Hanoteau

Parkinson

Tatham

(2010). Perceptions about memory reliability and honesty for children of 3 to 18 years old. Legal and Criminological Psychology, 15, 195-207. doi:10.1348/135532508X400347