You are on page 1of 13

Behavioral Sciences and the Law

Behav. Sci. Law 34: 126138 (2016)


Published online 26 February 2016 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/bsl.2211

Adults Insensitivity to Developmental Changes


in Childrens Ability to Report When and How
Many Times Abuse Occurred
Kyndra C. Cleveland and Jodi A. Quas*

In legal settings, children are frequently asked to provide temporal information about
alleged abuse, such as when it occurred and how often. Although there is a sizeable
body of work in the literature regarding childrens ability to provide such information,
virtually nothing is known about how adults evaluate the veracity of that information.
This omission is especially noteworthy given that adults evaluations are critical to
the progression and outcome of legal cases. We examined adults perceptions of
childrens reports of temporal details regarding alleged sexual abuse. We varied both
childrens age (6 vs. 11 years) and how certain children were when providing such
details to assess whether adults were sensitive to changes in how children of different
ages typically talk about temporal information. With regard to credibility, adults were
insensitive to childrens age, perceiving younger and older children who reported
temporal details with condence as more credible than those who reported information
tentatively. Normative developmental trends, however, would suggest that, with age,
children are often tentative when reporting true temporal details. With regard to
perceptions of childrens accuracy in reporting temporal information, adults found
younger children who were condent to be the most accurate. Regarding guilt
judgments, adults rated defendants as having a higher degree of guilt when children
were condent in reporting temporal details. The ndings have implications for juror
decision-making in cases of alleged sexual abuse in which children report when or
how often abuse occurred. Copyright # 2016 John Wiley & Sons, Ltd.

In child sexual abuse investigations, statements made by children are often given a
great deal of weight. Other forms of corroborating evidence may be scarce or nonexis-
tent, requiring that children provide considerable detail about the alleged abuse, such
as by whom, when, and how many times it occurred (Lamb & Brown, 2006). When
questions are not falsely suggestive or misleading, even relatively young children can
provide a wealth of accurate information, including details about a perpetrator, the lo-
cation of the abusive events, and other contextual details (see Malloy & Quas, 2009;
Saywitz, Lyon, & Goodman, 2011). Despite this sometimes impressive performance,
one type of detail remains particularly difcult for children to provide, namely temporal
information, such as when (e.g., what month) the abuse occurred, how old a child was
when the abuse started or ended, and how many times the abuse occurred. Even when
children are recounting a salient, documented past event, including one that occurred

*Correspondence to: Dr. Jodi Quas, Department of Psychology and Social Behavior, University of California,
Irvine, CA 92697-7085. E-mail: jquas@uci.edu

University of California, Irvine, CA

Copyright # 2016 John Wiley & Sons, Ltd.


Adults Perceptions of childrens Temporal Reports of Abuse 127

repeatedly, and even when other information they provide is accurate, childrens ability
to provide accurate temporal details is limited (Peterson, 1996; Wandrey, Lyon, Quas,
& Friedman, 2012).
Adults involved in legal cases and in other investigative contexts are in a challenging
position. They must evaluate the veracity of what children say about past events and
make decisions based on those evaluations, decisions that have signicant implications
for children, families, and the accused. Given that questions about temporal details reg-
ularly arise in investigations and trials (Orbach & Lamb, 2007; Powell, Roberts, &
Guadagno, 2007; R. v. R.W., 2006; U.S. v. Tsinhnahijinnie, 1997), it is important to
ascertain how adults evaluate childrens responses, including whether adults take chil-
drens limited temporal abilities into account when rendering judgments. We explored
these issues in the present study. We begin by providing an overview of childrens ac-
tual temporal abilities and how such abilities change with age. We then describe extant
research concerning adults (most often mock jurors) perceptions of child witnesses,
paying particular attention to how child witnesses age and condence inuence adults
perceptions. Thereafter we present the studys hypotheses and describe the study.

Childrens Temporal Abilities

A small but impressive set of studies has delineated important developmental changes
in childrens temporal memory and reporting. In most of these studies, children were
exposed to some event (contrived or naturally occurring) that occurred either once or
multiple times. At some later date, children were asked temporal questions about the
event(s), such as the day, month, and season in which the event(s) occurred (Friedman,
1991, 1992; Friedman & Lyon, 2005), or whether the event occurred before or after a
salient landmark event (e.g., Halloween). Friedman (1991), for example, interviewed
4-, 6-, and 8-year-olds about two classroom events, one that had taken place 7 weeks
previously and another that had taken place a week previously. Children were asked
whether they recalled the events (all did) and then when the events occurred, according
to four timescales: day of the week, time of day, month, and season. Children were also
asked to make judgments about the relative recency of the two events in comparison to
each other.
Friedmans results, which parallel those obtained in similar research, revealed con-
siderable decits in childrens abilities but also some developmental improvements.
The youngest children, for instance, were quite inaccurate regardless of the timescale,
and more than 50% of the time they reported that the events at school had occurred
during weekend days. The two older age groups performed better, or at least above
chance, when reporting season, month, and time of day, but not day of the week.
Across the age range, children were somewhat more accurate when judging the relative
recency of the two events. At least 70% of children in each age group correctly reported
which event came rst. The generalizability of this level of accuracy, though, is limited
given that one of the events had taken place particularly close in time to the interview
(see also Pathman, Larkina, Burch, & Bauer, 2013).
A few studies have examined childrens ability to report temporal details about
repetitive, emotionally charged events that occurred in the distant past (Peterson,
1996; Wandrey et al., 2012), events that are likely more salient to children than the
school-based play activities that have served as the to-be-remembered events in a
majority of the developmental studies. The results of these studies still revealed fairly

Copyright # 2016 John Wiley & Sons, Ltd. Behav. Sci. Law 34: 126138 (2016)
DOI: 10.1002/bsl
128 K. C. Cleveland and J. A. Quas

limited abilities. Wandrey et al. (2012) questioned 6- to10-year-old maltreated children


about when (e.g., time of day, month) and how many times (i.e., frequency) they had
experienced prior changes in placement (e.g., moving from foster family to foster fam-
ily) and when and how many times they had been to dependency court hearings previ-
ously. Both events are potentially salient and arousing for children and hence may be
well remembered (Davis, Quas, & Levine, 2008; Goodman et al., in press). In terms
of actual experiences, the children had, on average, changed placements three times
previously and had been to court more than three times. In addition, the delay between
the most recent of both of these events and the interview ranged from about 0.5 to
1.5 years ago, and were thus much longer than the delays in most experimental studies
of childrens temporal abilities.
In terms of how well children answered temporal questions, regardless of age, they
performed remarkably poorly, even though they typically provided other evidence indi-
cating that they remembered the events in question. Their poor performance was prob-
ably due to the longer delays between the documented events and the interview than was
the case in laboratory studies, although the nature of the sample (i.e., maltreated
children, who tend to be delayed cognitively relative to their non-maltreated counter-
parts; Kendall-Tackett & Eckenrode, 1996; Myers et al., 1999; Veltman & Browne,
2001) might also have led to some decreases in capabilities. Wandrey et al. (2012), for
instance, found that not one of the 6-year-olds accurately reported the month of their
most recent court visit, whereas children of comparable age in Friedmans studies could
report the month of the prior classroom events. Wandrey et al. also found that, across
age, only about 10% of the children accurately reported the exact month of either their
most recent prior placement or court visit, although many of the children reported accu-
rately who they had been placed with, suggesting that the children had some memory
and were willing to talk about the targeted events. Developmental improvements
emerged to some extent, with older children reporting certain temporal information at
levels above chance, but even here, performance was not especially impressive. Finally,
across questions, 1225% of the children simply said I dont know, an answer that
probably reects their honest lack of memory for temporal details.
The evident difculty children have recounting precise temporal information is not
surprising, given their more limited inferential and meta-memory abilities. With age,
individuals inferential abilities improve considerably. Thus, individuals can increas-
ingly rely on contextual information to infer temporal details. Indeed, adults often ap-
proximate temporal details, using such strategies as estimation or other temporal cues
to help them answer event frequency questions (see Friedman, 2004; Hasher & Zacks,
1984). Individuals metacognitive understanding of their own abilities also improves
substantially with age (Veenman, Van Hout-Wolters, & Aferbach, 2006). Thus, indi-
viduals increasingly understand their own mnemonic limitations, which may result in
greater recognition that they cannot remember all details about a prior event, including
possibly temporal details. This could lead to greater, and appropriate, use of uncer-
tainty terms or hedge language when answering at least some types of questions.
Hints at these possibilities have emerged in a few studies of childrens event
reporting and responses to temporal questions. Wandrey (2013), for example, coded
hundreds of trial transcripts of children testifying in sexual abuse cases for the inclusion
of temporal information (e.g., how many times and when abuse happened). With age,
children were more likely to answer temporal questions, but also to include phrases
indicative of uncertainty in their responses. Similarly, studies of childrens responses

Copyright # 2016 John Wiley & Sons, Ltd. Behav. Sci. Law 34: 126138 (2016)
DOI: 10.1002/bsl
Adults Perceptions of childrens Temporal Reports of Abuse 129

to nonsense questions and studies of the effects of forensic interviewing instructions on


childrens performance have revealed that, with age, children naturally and without
prompting increasingly provide dont know responses, including in situations for which
dont know is the correct answer because there is no accurate answer (Earhart,
La Rooy, Brubacher, & Lamb, 2014; Waterman & Blades, 2011; Waterman, Blades,
& Spencer, 2000). As we discuss next, these responses, which are indicative of
uncertainty and may well be developmentally appropriate and accurate depictions of
childrens honest lack of knowledge, have implications for how others evaluate
childrens credibility.

Adults Perceptions of Child Witnesses

Although previous studies have failed to examine how adults evaluate temporal details
provided by children, studies have examined adults general perceptions of childrens
memory, suggestibility, and eyewitness capabilities (Goodman, Golding, & Haith,
1984; Leippe, Manion, & Romanczyk, 1992; Nunez, Kehn, & Wright, 2011; Ross,
Dunning, Toglia, & Ceci, 1990; Schmidt & Brigham, 1996). Two characteristics of
children that shape adults perceptions and that may be particularly important when
adults evaluate childrens reports of temporal information are the childrens age and
level of condence when responding.
With regard to age, across experimental designs in which adults read transcripts of
childrens responses or view videotaped mock criminal trials, adults including college
students, community members, and actual jurors perceive younger children (e.g.,
preschool and young school age) as less credible than older children (e.g., middle
school age) and adults [Goodman et al., 1984, 1987; Leippe et al., 1992; Leippe &
Romanczyk, 1987, 1989 (experiment 2); Schmidt & Brigham, 1996]. Adults tend to
believe that younger children have poorer memory in general and can be easily manip-
ulated by others into reporting false information (Goodman et al., 1984; Leippe &
Romanczyk, 1987, 1989). Of note, although adults perceptions of childrens credibil-
ity increase with child age, this increase slows as children transition to the teen years,
with adolescents at times being perceived as less credible than middle school-age chil-
dren, a trend that seems to reect adults beliefs about the possibility of adolescents
lying rather than limited cognitive abilities (e.g., Duggan et al., 1989; Goodman,
Bottoms, Herscovici, & Shaver, 1989).
Insofar as adults believe that memory improves and suggestibility decreases with age,
at least through middle childhood, adults would probably believe temporal memory
follows a similar pattern. Thus, their evaluations about age-related increases in accu-
racy and credibility should extend to situations in which children provide temporal
details.
As mentioned, however, with age, childrens meta-memory abilities improve, which
could lead to age-related increases in childrens willingness to admit a lack of knowl-
edge and, as a result, demonstrate more expressions of uncertainty when answering
temporal questions (Wandrey, 2013). Such expressions could affect adults percep-
tions, both directly and in conjunction with age. Research on witness condence,
although not in relation to temporal reporting, offers some hints. In juror decision-
making studies, witness condence has typically been manipulated by varying how
often witnesses include statements of uncertainty or behave in a manner indicative of
high versus low power. Not surprisingly, condent adult witnesses are evaluated more

Copyright # 2016 John Wiley & Sons, Ltd. Behav. Sci. Law 34: 126138 (2016)
DOI: 10.1002/bsl
130 K. C. Cleveland and J. A. Quas

positively than uncondent adult witnesses (Whitley & Greenberg, 1986), which corre-
spondingly often leads to increases in guilty verdicts (Brewer & Burke, 2002). Jurors
positive evaluation of witness condence extends to cases in which children are wit-
nesses (Goodman et al., 1984).
In one study, for example, Schmidt and Brigham (1996) had mock jurors rate the
credibility of a 5-, 10-, or 15-year-old girl after watching a video of her testifying in a
sexual abuse trial. The girls condence (termed power in the original study) was var-
ied, such that she responded to attorney questions in either a low-condence manner,
for example, by dgeting and including hedge words (huh and um), or a high-
condence manner, that is, by speaking clearly with no uncertainty. Jurors who viewed
the low-condence girl rated her less positively than jurors who viewed the high-
condence girl. This effect was further moderated by age. Condence had a greater
effect on evaluations of the 5- and 15-year-old than of the 10-year-old, with high con-
dence enhancing positive evaluations of the 5-year-old and low condence decreasing
positive evaluations of the 15-year-old. Schmidt and Brigham (1996) speculated that
jurors might have expected younger children to exhibit less condence in their speech
generally, and therefore viewed the young condent child as highly credible because
her performance exceeded expectations. At the same time, the jurors may have
expected the 15-year-old to display condence and thus reacted negatively when she
exhibited low condence. The latter interpretation (but not the former) ts with other
studies that have found that jurors tend to evaluate witnesses more negatively when
they behave inconsistently with jurors expectations, a phenomenon often referred to
as expectancy violation (Hackett, Day, & Mohr, 2008; Kaufmann et al., 2003).
Insofar as children increasingly recognize the limitations of their memory reports, es-
pecially in terms of remembering temporal details, with age they may naturally include
higher numbers of expressions of uncertainty when responding to temporal questions
(Wandrey, 2013). Adults may not recognize the developmental appropriateness of
older childrens uncertain answers and evaluate older uncertain children more nega-
tively. Whether adults view young children who are certain as especially credible, sim-
ilar to Schmidt and Brighams ndings, or view young children who are certain with
skepticism because they should not be certain is unclear from prior research. In light
of evidence that temporal questions are routinely asked of children and routinely
answered, how adults evaluate childrens certain and uncertain responses is of
considerable interest.

Study Overview and Hypotheses

In the current study, adults read a trial transcript, presented online, of a female child
testifying in a sexual abuse case. The age of the child witness (6 vs. 11 years) and her
condence in reporting temporal information (sure vs. unsure) varied. Afterwards,
adults rated the credibility of the child and the defendant, and provided a case verdict
and guilt rating for the defendant. We hypothesized a signicant main effect of tempo-
ral condence, such that adults would rate children who were sure in response to tem-
poral questions as more credible than children who were unsure; and a signicant
interaction between age and temporal condence, such that in the sure condition,
adults would rate older and younger children as high in credibility, whereas in the un-
sure condition, adults would rate older children as less credible. In contrast to Schmidt
and Brighams (1996) results in which highly certain young children were evaluated

Copyright # 2016 John Wiley & Sons, Ltd. Behav. Sci. Law 34: 126138 (2016)
DOI: 10.1002/bsl
Adults Perceptions of childrens Temporal Reports of Abuse 131

particularly favorably, we did not expect younger childrens credibility to be particularly


high in the sure condition. Our prediction was based on studies showing that adults
generally hold lower expectations about younger childrens capabilities, and sure testi-
mony from a young child violates those expectations. Finally, with regard to verdicts
and degree of guilt ascribed to the defendant, we predicted similar main effects and in-
teractions, namely that adults in the sure child witness condition would be more likely
to render guilty verdicts and rate the defendant as having a higher degree of guilt than
adults in the unsure child witness condition, and that the lower degree of guilt in the
unsure condition would be most notable for older children.

METHOD
Participants

A total of 173 psychology undergraduate students participated in the current study


(54.9% female) in exchange for course extra credit. The inclusion criterion was that
participants should be eligible for jury duty 18 years of age or older and a U.S. citizen
(Mage = 21.02, SD = 4.11). Ethnicity was consistent with the demographics of the uni-
versity population: 52% Asian-American, 19.1% White, 24.9% Hispanic/Latino,
1.7% African-American, 0.6% American-Indian, 9.8% Other.1 Four students were
excluded because they were not U.S. citizens or they did not indicate their citizenship
status.

Procedure

All procedures were approved by the University of California, Irvine Institutional


Review Board. Participants provided consent before beginning the online survey,
which took approximately 1 hour to complete. The survey began with a brief summary
of a child sexual abuse case, modeled after an actual case, followed by a mock trial
transcript including the childs testimony during direct and cross-examination and
re-direct and re-cross-examination, presented in a format similar to that in prior studies
(e.g., Cooper, Quas, & Cleveland, 2014).
After reading the transcript, participants were asked to provide a verdict and rate the
defendants guilt on a scale ranging from 1 (denitely not guilty) to 6 (denitely guilty).
Next, participants were asked to rate the child and defendant on several dimensions
using a six-point Likert scale from 1 (not at all) to 6 (very). Questions about the child
included one set of items that asked about general characteristics or behaviors of the
child, such as her intelligence, honesty, believability, general memory, and likeability;
and another set asked specically about the childs performance when recounting tem-
poral information. These concerned how accurately the child recalled when the abuse
occurred, how well she reported when it occurred, how accurately she recalled how
many times the abuse occurred, and how well she reported the frequency of abusive in-
cidents. Questions about the defendant paralleled those about the general characteris-
tics of the child, asking about his intelligence, honesty, believability, and likeability.

1
Total percentage is greater than 100% because participants were asked to choose all that apply, and thus
they may be included in more than one category.

Copyright # 2016 John Wiley & Sons, Ltd. Behav. Sci. Law 34: 126138 (2016)
DOI: 10.1002/bsl
132 K. C. Cleveland and J. A. Quas

Finally, participants were asked about general perceptions of child witnesses (e.g.,
whether they thought inconsistencies in a childs report of sexual abuse were indicative
of the report being false) on a six point scale, ranging from strongly disagree to strongly
agree, and about basic demographic information (e.g., age, gender, ethnicity, marital
status, salary, experience and time spent with children, citizenship status, and prior
legal experience as a victim or suspect).

Trial Transcript

The case involved child sexual abuse in which a girl was allegedly abused, on several
occasions over the course of the previous year, by her father. The childs mother was
divorced from the childs father, and immediately contacted authorities upon hearing
about the alleged abuse from her daughter. The childs father denied the allegations
against him, and his employer and friend testied in support of his claims. The trial
transcript provided the testimony of the child complainant answering demographic
questions and questions about the alleged abuse, which included temporal questions
regarding when and how often the abuse occurred.

Manipulation

The childs temporal condence (sure vs. unsure) and the childs age (6 vs. 11 years)
were manipulated across transcripts. In the sure condition, the child answered tempo-
ral questions with condence (e.g., attorney question: Are you sure it was March?;
child answer: Yes, Im sure.). In the unsure condition, uncertainty terms were em-
bedded within the childs answers to the temporal questions (e.g., attorney question:
Are you sure it was winter?; child answer: I think so or I dont know.). Across
transcripts, questions included those that asked about temporal location (e.g., What
month was it?) and numerosity (e.g., Did he touch you more than ve times?).

RESULTS
Data Reduction and Preliminary Analyses

Participants Likert ratings regarding the child and defendant were combined to create
composite scores, similar to that done in prior studies (e.g., Myers et al., 1999; Redlich,
Ghetti, & Quas, 2008). One composite reected participants perceptions of the childs
generally credibility, created by averaging their answers (which were signicantly corre-
lated: r-values ranged from 0.27 to 0.83, p-values < 0.001) about characteristics
reecting the childs honesty, overall memory accuracy, likeability, etc. ( = 0.95). Be-
cause of our specic interest in participants perceptions of the childs temporal abili-
ties, participants ratings to items about the childs temporal performance (r-values
ranged from 0.56 to 0.85, p-values < 0.001) were averaged separately to create a tem-
poral accuracy score ( = 0.91). The two sets of scores were strongly correlated
(r = 0.84, p < 0.001), but led to slightly different patterns of results and hence were an-
alyzed separately. Participants Likert scale ratings of the defendant (r-values ranged
from 0.27 to 0.74, p-values < 0.001, = 0.81) were similarly combined to create a com-
posite defendant credibility score.

Copyright # 2016 John Wiley & Sons, Ltd. Behav. Sci. Law 34: 126138 (2016)
DOI: 10.1002/bsl
Adults Perceptions of childrens Temporal Reports of Abuse 133

Preliminary analyses tested for potential confounds. Correlations between partici-


pants age and their perceptions of the childs general credibility and temporal accuracy
and defendant guilt ratings and verdicts revealed that, with age, participants ratings of
the childs general credibility and temporal accuracy increased (r-values 0.166, p-
values 0.028). When participant ethnicity (White, Asian American, Hispanic/Latino,
and Other) was examined via a one-way analysis of variance (ANOVA), post hoc Scheffe
tests revealed that White participants rated the child as more credible than did partici-
pants who reported they were only Hispanic/Latino and as more temporally accurate
than participants who reported they were only some other race/ethnicity [all F(3,
166) 3.15, p-values 0.027, 2p = 0.05]. Whether participants had children, how much
experience they had with children, and how much time they spent with children were
not signicantly related to the dependent variables and thus are not considered further.
We return to the issues of participant age and ethnicity in the Discussion section.

Main Analyses

Child General Credibility and Temporal Accuracy

Two temporal condence (sure vs. unsure) child age (6 vs. 11 years) participant
gender (female vs. male) ANOVAs were conducted, one predicting participants per-
ceptions of childrens general credibility and the other predicting participants percep-
tions of childrens temporal accuracy. All main effects and interactions were tested. For
general credibility, only the main effect of temporal condence was signicant [F(1,
169) = 33.71, p < 0.001, 2p = 0.17]. Participants rated children in the sure condition
(M = 3.57, SD = 1.05) as substantially more credible than children in the unsure condi-
tion (M = 2.72, SD = 0.81).
When participants ratings of the childs temporal accuracy were considered, the
main effect of temporal condence and the age condence interaction were both sig-
nicant [all F(1, 163) 4.35, p-values 0.039, 2p 0.026]. Again, robust differences
were evident when comparing ratings of children in the sure versus unsure conditions
(M = 3.10, SD = 0.17, and M =1.94, SD = 0.73, respectively). Planned comparisons re-
vealed that, when children were unsure, no signicant age differences emerged in par-
ticipants ratings of childrens temporal accuracy. However, younger childrens sure
temporal reports were rated as slightly more accurate than older childrens sure reports
[t(163) = 1.77, p < 0.10], and signicantly more accurate than older childrens unsure
reports [t(163) = 6.11, p < 0.001] (Figure 1).
Finally, when participants ratings of defendant credibility were considered, no sig-
nicant effects emerged. Thus, although the certainty with which children provided
temporal details inuenced participants perceptions of the childrens general credibil-
ity and temporal accuracy, this did not spill over and alter how the participants viewed
the defendants credibility.

Guilt Preferences

We anticipated that not only would unsure children be evaluated more negatively, but
that in the unsure condition, there would also be a reduction in participants percep-
tions of defendant guilt. We tested this possibility by examining participants verdict
judgments and their Likert scale guilt ratings. For verdict preferences (0 not guilty,

Copyright # 2016 John Wiley & Sons, Ltd. Behav. Sci. Law 34: 126138 (2016)
DOI: 10.1002/bsl
134 K. C. Cleveland and J. A. Quas

Figure 1. Adults perceptions of childrens temporal accuracy by age and condence. Ratings refer to adults
preceptions of childrens performance when recounting temporal information on a scale ranging from 1 (not
accurate at all) to 6 (very accurate).

1 guilty), a logistic regression was conducted with temporal condence, child age,
and participant gender as independent variables. No signicant effects emerged
[ 2(3) = 4.36, p = 0.23]. When participants guilt ratings were entered into a temporal
condence child age participant gender ANOVA, the main effect of temporal con-
dence was signicant [F(1, 165) = 4.70, p = 0.032, 2p = 0.03]. Participants rated the
defendant as having a signicantly lower degree of guilt when the child witness was un-
sure (M = 3.35, SD = 1.08) versus sure (M = 3.77, SD = 1.15) in response to temporal
questions.

DISCUSSION
Adults, including professionals ranging from teachers and nurses to social workers, po-
lice, and jurors, are frequently in a position where they must question children about an
alleged event and evaluate the veracity and completeness of childrens responses. Ju-
rors judgments, in particular, are linked to decisions about guilt, decisions that affect
the lives of children, families, and the accused. In such cases, temporal information
(e.g., when an incident happened, how many times) may be critical, particularly with
regard to conrming or disconrming alibies and determining the number of charges
to be brought against the defendant. Children can, at times, provide some temporal in-
formation, but their abilities are limited, at least through middle childhood (Friedman,
1986, 1991; Friedman & Lyon, 2005). Adults seem unaware of these limitations and
seem to assume temporal inaccuracies or uncertainty when reporting temporal
information is reective of lower accuracy more generally.
In other words, in our study, jury-eligible adults appeared to penalize children for
being unsure about dates and frequencies of alleged abuse experienced in the distant
past, rendering lower likelihood of guilt ratings than when children were sure. Although
signicant results regarding verdicts (guilty or not) did not emerge, adults negative
evaluations of childrens uncertain statements are nonetheless signicant precisely

Copyright # 2016 John Wiley & Sons, Ltd. Behav. Sci. Law 34: 126138 (2016)
DOI: 10.1002/bsl
Adults Perceptions of childrens Temporal Reports of Abuse 135

because they demonstrate adults lack of knowledge about childrens development and
likely capabilities. More specically, the adults perceptions in our study are at odds with
empirical research on childrens actual temporal abilities, their developing metacognitive
skills, and the way in which children describe temporal information (Friedman, 1991;
Wandrey, 2013; Waterman & Blades, 2011). With increasing age, children may be more
aware that they do not know the answers to some temporal questions and may be more
comfortable reporting their lack of knowledge, or at least acknowledging uncertainty.
By doing so, though, children run the risk of undermining how credible they are perceived
to be by adults, leading to more honest, but more negatively viewed, reports.
We were somewhat surprised, in light of differences in how adults generally evaluate
childrens reports, and in light of some studies that have revealed differences in how adults
view certain versus uncertain responses from children who vary with age, that more robust
age differences were not evident, particularly in conjunction with certainty. On the one
hand, perhaps our age groups were not sufciently divergent. Had we included adoles-
cents, who, in sexual abuse cases, are often perceived as less credible relative to middle
school-age children, we may have uncovered larger (or instead more varying) age trends.
On the other hand, some explanations may account for the lack of age differences here.
One is, quite simply, that the adults did not take into account childrens developing abil-
ities overall. If such were the case, jurors should have rated the older children as more
credible than the younger children. This should have existed regardless of child certainty,
although it should have been more pronounced in the sure condition if adults had relied
on the type of beliefs evident in other studies, namely adults believe that, with increasing
child age, memory improves and suggestibility decreases. A second possibility is that
adults specically failed to recognize age-related increases in childrens likely use of un-
certainty terms when reporting some types of temporal information. This then led the
adults to simply discount all uncertain responses, regardless of age (see Brewer & Burke,
2002, for ndings showing that witness condence overrides other factors in affecting per-
ceptions and judgments of guilt). Of interest, even though adults often view younger chil-
dren as having poorer memory (e.g., Goodman et al., 1984), adults did not, in the current
study, evaluate uncertain responses from younger children as any more positive, or per-
haps as more acceptable, than uncertain responses from older children. Instead, chil-
drens lack of condence, across age, simply led to lower perceived general credibility,
and also lower perceptions of accuracy in recounting temporal details. In contrast, when
young children were certain, their temporal reports were considered the most accurate,
somewhat more so than similarly certain reports provided by older children, but especially
more so than older childrens uncertain reports. These trends are similar to Schmidt and
Brighams ndings, who also found particularly high ratings of perceived credibility when
adults viewed young children answering with high levels of condence.
The lack of broad or consistent differences in perceptions of the younger and older
children is especially interesting, and highlights the need for better education for adults
about childrens emerging mnemonic (including temporal) abilities. Children, through
at least middle childhood, have difculty remembering (or reporting precisely) when
events happened and how often events occurred. Yet they are called upon to provide this
information in a variety of settings. Adults may hold unrealistic expectations about how
well children can provide these details, at least as evidenced by lower credibility ratings
when children express uncertainty. Whether education would help adults recognize ap-
propriate age-related limitations and also differences in how children report temporal in-
formation is an important question in need of addressing. It would also be worthwhile to

Copyright # 2016 John Wiley & Sons, Ltd. Behav. Sci. Law 34: 126138 (2016)
DOI: 10.1002/bsl
136 K. C. Cleveland and J. A. Quas

ascertain how adults evaluate variations in childrens certainty across different types of
questions.
As a nal note, although this study is unique in its focus on how adults evaluate chil-
drens reporting of temporal information, transcript-based studies of adults perceptions,
particularly those that include college-age samples, are limited. Transcripts do not
provide information about childrens demeanor or tone when reporting uncertain or
certain answers, and witness demeanor does affect adults evaluations (Cooper et al.,
2014; Golding, Fryman, Marsil, & Yozwiak, 2003; McAuliff & Bull Kovera, 2012). It will
be important to extend this work to include videotaped interviews or statements provided
by children. Also, although we varied childrens temporal statement certainty in a manner
consistent with what developmental science would predict children say, future studies
should present actual statements made by children and assess how adults evaluate those
statements. Finally, we relied on a convenience sample of college students, who tend
not to match demographically (e.g., age, ethnicity) eligible juror samples. Studies that
have compared the two groups often fail to nd consistent differences, and when differ-
ences emerge, they tend to be small (e.g., Buck, Warren, Bruck, & Kuehnle, 2014; Quas
et al., 2005; see Bornstein, 1999; Reed & Bornstein, 2015). Nonetheless, there were hints
that, as respondent age increased, perceptions of childrens credibility increased (across
experimental conditions), and White participants rated the child as slightly more credible
than Hispanic/Latino participants and more temporally accurate than participants who
reported themselves to be some other ethnicity or race. Although we are unsure of the spe-
cic reasons why the latter differences emerged, particularly in light of prior studies which
have typically not found ethnic/racial differences in perceptions of child credibility (e.g.,
Holcomb & Jacquin, 2007), the generalizability of the ndings here should be replicated
with other, diverse samples. These samples should include not only individuals whose
age, gender, and ethnicity match the demographics of actual jury pools, but also other
professionals (e.g., forensic interviewers, police ofcers, attorneys, teachers) who are of-
ten in positions of needing to question children about past events.
In closing, these results suggest that adults may not be aware of childrens temporal
abilities, or rather of developmental changes in childrens likely use of uncertainty when
describing temporal information about past events. Childrens use of uncertain lan-
guage, such as verbal hedges and dont know responses, may actually reect greater
honesty, especially for older children. And yet, adults seem unaware of this, viewing
childrens temporal responses as much more accurate when children used certain ver-
sus uncertain language. Adults, therefore, need substantial education about childrens
capabilities when recalling temporal information so that adults can more accurately
judge childrens responses and make decisions based on those judgments. As long as
young children are asked to provide temporal information in court, research should
continue to investigate how childrens abilities to report temporal information inu-
ence jurors perceptions and case decisions. By educating adults about childrens
temporal abilities, especially age differences in those abilities, child witnesses can be
judged in a more fair and accurate manner, and justice can be better served.

ACKNOWLEDGEMENTS
We wish to thank Lindsay Wandrey for her work on earlier versions of this study and
her assistance with setting up data collection protocols.

Copyright # 2016 John Wiley & Sons, Ltd. Behav. Sci. Law 34: 126138 (2016)
DOI: 10.1002/bsl
Adults Perceptions of childrens Temporal Reports of Abuse 137

REFERENCES

Bornstein, B. H. (1999). The ecological validity of jury simulations: Is the jury still out? Law and Human
Behavior, 23, 7591. doi:10.1023/A:1022326807441.
Brewer, N., & Burke, A. (2002). Effects of testimonial inconsistencies and eyewitness condence on mock-
juror judgments. Law and Human Behavior, 26, 353364. doi:10.1023/A:1015380522722.
Buck, J. A., Warren, A. R., Bruck, M., & Kuehnle, K. (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, 867883. doi:10.1002/bsl.2150.
Cooper, A., Quas, J. A., & Cleveland, K. C. (2014). The emotional child witness: Effects on juror decision
making. Behavioral Sciences & the Law, 32, 813828. doi:10.1002/bsl.2153.
Davis, E., Quas, J. A., & Levine, L. J. (2008). Childrens memory for stressful events: Exploring the role of
discrete emotions. In M. Howe, D. Cicchetti, & G. S. Goodman (Eds.). Stress, trauma, and childrens
memory development: Neurobiological, cognitive, clinical, and legal perspectives (pp. 236264). Oxford, UK:
Oxford University Press.
Duggan, L. M., Aubrey, M., Doherty, E., Isquith, P., Levine, M., & Scheiner, J. (1989). The credibility of
children as witnesses in a simulated child sex abuse trial. In S. J. Ceci, D. F. Ross, & M. P. Toglia
(Eds.). Perspectives on childrens testimony (pp. 7199). New York, NY: Springer.
Earhart, B., La Rooy, D. J., Brubacher, S. P., & Lamb, M. E. (2014). An examination of dont know responses
in forensic interviews with children. Behavioral Sciences & the Law, 32, 746761. doi:10.1002/bsl.2141.
Friedman, W. J. (1986). The development of childrens knowledge of temporal structure. Child Development,
57, 13861400. doi:10.2307/1130418.
Friedman, W. J. (1991). The development of childrens memory for the time of past events. Child
Development, 62, 139155. doi:10.1111/j.1467-8624.1991.tb01520.x.
Friedman, W. J. (1992). Childrens time memory: The development of a differentiated past. Cognitive
Development, 7, 171187. doi:10.1016/0885-2014(92)90010-O.
Friedman, W. J. (2004). Time in autobiographical memory. Social Cognition, 22, 591605. doi:10.1521/
soco.22.5.591.50766.
Friedman, W. J., & Lyon, T. D. (2005). Development of temporal-reconstructive abilities. Child Development,
76, 12021216.
Golding, J. M., Fryman, H. M., Marsil, D. F., & Yozwiak, J. A. (2003). Big girls dont cry: The effect of child
witness demeanor on juror decisions in a child sexual abuse trial. Child Abuse & Neglect, 27, 13111321.
doi:10.1016/j.chiabu.2003.03.001.
Goodman, G. S., Bottoms, B. L., Herscovici, B. B., & Shaver, P. R. (1989). Determinants of the child
victims perceived credibility. In S. J. Ceci, D. F. Ross, & M. P. Toglia (Eds.). Perspectives on childrens
testimony (pp. 122). New York, NY: Springer.
Goodman, G. S., Goldfarb, D., Quas, J. A., Narr, R., Milojevich, H., & Cordon, I. (in press). Memory
development, emotion regulation, and trauma-related psychopathology. To appear in D. Cicchetti (Ed.).
Developmental psychopathology: Risk, disorder, and adaptation. New York, NY: Wiley.
Goodman, G. S., Golding, J. M., & Haith, M. M. (1984). Jurors reactions to child witnesses. Journal of Social
Issues, 40, 139156. doi:10.1111/j.1540-4560.1984.tb01098.x.
Goodman, G. S., Golding, J. M., Helgeson, V. S., Haith, M. M., & Michelli, J. (1987). When a child takes
the stand: Jurors perceptions of childrens eyewitness testimony. Law and Human Behavior, 11, 2740.
doi:10.1007/BF01044837.
Hackett, L., Day, A., & Mohr, P. (2008). Expectancy violation and perceptions of rape victim credibility.
Legal and Criminological Psychology, 13, 323334. doi:10.1348/135532507X228458.
Hasher, L., & Zacks, R. T. (1984). Automatic processing of fundamental information: The case of frequency
of occurrence. American Psychologist, 39, 13721388. doi:10.1037/0003-066X.39.12.1372.
Holcomb, M. J., & Jacquin, K. M. (2007). Juror perceptions of child eyewitness testimony in a sexual abuse
trial. Journal of Child Sexual Abuse, 16, 7995. doi:10.1300/J070v16n02_05.
Kaufmann, G., Drevland, G. C. B., Wessel, E., Overskeid, G., & Magnussen, S. (2003). The importance of
being earnest: Displayed emotions and witness credibility. Applied Cognitive Psychology, 17, 2134.
doi:10.1002/acp.842.
Kendall-Tackett, K. A., & Eckenrode, J. (1996). The effects of neglect on academic achievement and
disciplinary problems: A developmental perspective. Child Abuse & Neglect, 20, 161169. doi:10.1016/
S0145-2134(95)00139-5.
Lamb, M. E., & Brown, D. (2006). Conversational apprentices: Helping children become competent
informants about their own experiences. British Journal of Developmental Psychology, 24, 215234.
doi:10.1348/026151005X57657.
Leippe, M. R., Manion, A. P., & Romanczyk, A. (1992). Eyewitness persuasion: How and how well do fact
nders judge the accuracy of adults and childrens memory reports? Journal of Personality and Social
Psychology, 63, 181197. doi:10.1037/0022-3514.63.2.181.

Copyright # 2016 John Wiley & Sons, Ltd. Behav. Sci. Law 34: 126138 (2016)
DOI: 10.1002/bsl
138 K. C. Cleveland and J. A. Quas

Leippe, M. R., & Romanczyk, A. (1987). Children on the witness stand: A communication/ persuasion
analysis of jurors reactions to child witnesses. In S. J. Ceci, M. P. Toglia, & D. F. Ross (Eds.). Childrens
eyewitness memory (pp. 155177). New York, NY: Springer.
Leippe, M. R., & Romanczyk, A. (1989). Reactions to child (versus adult) eyewitnesses: The inuence of
jurors preconceptions and witness behavior. Law and Human Behavior, 13, 103132. doi:10.1007/
BF01055919.
Malloy, L., & Quas, J. A. (2009). Childrens suggestibility. Areas of consensus and controversy. In K.
Kuehnle & M. Connoll (Eds.). Critical issues in child sexual abuse assessment (pp. 267297). Sarasota, FL:
Professional Resource Press.
McAuliff, B. D., & Bull Kovera, M. (2012). Do jurors get what they expect? Traditional versus alternative
forms of childrens testimony. Psychology, Crime & Law, 18, 2747. doi:10.1080/1068316X.2011.613391.
Myers, J. E. B., Berliner, L., Briere, J., Hendrix, C. T., Jenny, C., & Reid, T. (1999). The APSCAC handbook
on child maltreatment. Thousand Oaks, CA: Sage.
Myers, J. E. B., Redlich, A. D., Goodman, G. S., Prizmich, L. P., & Imwinkelried, E. (1999). Jurors
perceptions of hearsay in child sexual abuse cases. Psychology, Public Policy, & Law, 5, 388419.
doi:10.1037/1076-8971.5.2.388.
Nunez, N., Kehn, A., & Wright, D. B. (2011). When children are witnesses: The effects of context, age and
gender on adults perceptions of cognitive ability and honesty. Applied Cognitive Psychology, 25, 460468.
doi:10.1002/acp.1713.
Orbach, Y., & Lamb, M. E. (2007). Young childrens references to temporal attributes of allegedly
experienced events in the course of forensic interviews. Child Development, 78, 11001120.
Pathman, T., Larkina, M., Burch, M. M., & Bauer, P. J. (2013). Young childrens memory for the times of
personal past events. Journal of Cognition and Development, 14, 120140. doi:10.1080/15248372.2011.641185.
Peterson, C. (1996). The preschool child witness: Errors in accounts of traumatic injury. Canadian Journal of
Behavioural Science, 28, 3642. doi:10.1037/0008-400X.28.1.36.
Powell, M., Roberts, K., & Guadagno, B. (2007). Particularisation of child abuse offences: Common
problems when questioning child witnesses. Current Issues in Criminal Justice, 19, 6474.
Quas, J. A., Thompson, W. C., & Clarke-Stewart, K. A. (2005). Do jurors" know" what isnt so about child
witnesses? Law and Human Behavior, 29, 425456. doi:10.1007/s10979-005-5523-8.
R. v. R.W., No. CanLII 19283 (Can. Ont. C.A.) (2006).
Redlich, A. D., Ghetti, S., & Quas, J. A. (2008). Perceptions of children during a police interview: A
comparison of alleged victims and suspects. Journal of Applied Social Psychology, 38, 705735.
doi:10.1111/j.1559-1816.2007.00323.x.
Reed, K., & Bornstein, B. H. (2015). Using mock jury studies to measure community sentiment toward child
sexual abusers. In M. Miller, J. A. Blumenthal, & J. Chamberlain (Eds.). Handbook of community sentiment
(pp. 5768). New York, NY: Springer.
Ross, D. F., Dunning, D., Toglia, M. P., & Ceci, S. J. (1990). The child in the eyes of the jury: Assessing
mock jurors perceptions of the child witness. Law and Human Behavior, 14, 523. doi:10.1007/
BF01055786.
Saywitz, K. J., Lyon, T. D., & Goodman, G. S. (2011). Interviewing children. In J. E. B. Myers (Ed.). The
APSAC handbook on child maltreatment (3rd ed.), (pp. 337360). Newbury Park, CA: Sage.
Schmidt, C. W., & Brigham, J. C. (1996). Jurors perceptions of child victim-witnesses in a simulated sexual
abuse trial. Law and Human Behavior, 20, 581606. doi:10.1007/BF01499233.
U.S. v. Tsinhnahijinnie, 112 F. 3d 988 (9th Cir. 1997).
Veenman, M. V., Van Hout-Wolters, B. H., & Aferbach, P. (2006). Metacognition and learning:
Conceptual and methodological considerations. Metacognition and Learning, 1, 314. doi:10.1007/
s11409-006-6893-0.
Veltman, M. W. M., & Browne, K. D. (2001). Three decades of child maltreatment research: Implications
for the school years. Trauma, Violence & Abuse, 2, 215239. doi:10.1177/1524838001002003002.
Wandrey, L. (2013). How many times did he do this to you? Attorneys temporal questions during child
sexual abuse trials. (Doctoral dissertation). Retrieved from ProQuest. (Publication No. 3565436)
Wandrey, L., Lyon, T. D., Quas, J. A., & Friedman, W. J. (2012). Maltreated childrens ability to estimate
temporal location and numerosity of placement changes and court visits. Psychology, Public Policy, &
Law, 18, 79104. doi:10.1037/a0024812.
Waterman, A. H., & Blades, M. (2011). Helping children correctly say I dont know to unanswerable
questions. Journal of Experimental Psychology: Applied, 17, 396405. doi:10.1037/a0026150.
Waterman, A. H., Blades, M., & Spencer, C. (2000). Do children try to answer nonsensical questions? British
Journal of Developmental Psychology, 18, 211225. doi:10.1348/026151000165652.
Whitley, B. E., & Greenberg, M. S. (1986). The role of eyewitness condence in juror perceptions of
credibility. Journal of Applied Social Psychology, 16, 387409. doi:10.1111/j.1559-1816.1986.tb01148.x.

Copyright # 2016 John Wiley & Sons, Ltd. Behav. Sci. Law 34: 126138 (2016)
DOI: 10.1002/bsl