Forensic interview

Instructions.  Apply your knowledge of interviewing best practices to critique the example forensic interview.  In the hypothetical interview, the interviewer is attempting to better understand a 4-year-old child’s experiences.  Your job is to…
a. Identify 10 problem areas/challenges within the interview.  For each problem or challenge you must …
b. Explain why the interaction or interviewing technique is problematic from anempirical perspective (e.g., the use of closed ended questions are problematicbecause research shows…).  Support your explanation by using knowledgelearned in the course.  You should be as explicit as possible in explaining theproblems you identify.  Be sure to include APA style in-text citations and references.
c.  Tell me how you would work to fix the problem/overcome the challenge.  To earn full points, you must explain why your recommendation falls within best practice guidelines.  Support your ideas by citing relevant sources in APA style.
Divining testimony? The impact of interviewing props
on children’s reports of touching
Debra Ann Poole a,⇑
, Maggie Bruck b
aDepartment of Psychology, 231 Sloan Hall, Central Michigan University, Mt. Pleasant, MI 48859, United States
bDivision of Child and Adolescent Psychiatry, Johns Hopkins University, United States
article info
Article history:
Available online 4 July 2012
Forensic interviewing
Sexual abuse
Interviewing props
There is a long-held assumption that objects help bridge the gap
between what children know and what they can (or are willing
to) explain. In this review, we present research on the extent to
which two types of objects used as props in investigative interviews
of children, anatomical dolls and body (human figure) diagrams,
actually help children report accurate information about
autobiographical events. We explain why available research does
not instill confidence that props are the best solution to interviewing
challenges, and we consider practitioners’ and policy-makers
responses to this evidence. Finally, we discuss the types of developmental
research that are necessary to advance the field of evidence-based
interviewing of children.
2012 Elsevier Inc. All rights reserved.
Authorities have always had faith that special techniques would help them investigate crimes. For
thousands of years, divination with objects was a widespread phenomenon that persisted in some regions
despite frequent ridicule and bans on its use (e.g., Beard, North, & Price, 1998). In one form of the
practice, an inquisitor asked a young child to stare at an object, which presumably allowed only the
child to receive information from spirits about thieves or other pressing issues. To summon ‘‘the
princes of the oil,’’ adults prepared a shiny surface and then instructed the child to look for the beings
who revealed society’s secrets (Bilu, 1981).
Most forms of scrying (divination by gazing at objects) involved young boys or virgin girls because
their purity presumably gave them ‘‘unparalleled abilities of revelation and divination’’ (Bilu, 1981, p.
0273-2297/$ – see front matter 2012 Elsevier Inc. All rights reserved.
⇑ Corresponding author. Fax: +1 989 774 2553.
E-mail address: (D.A. Poole).
Developmental Review 32 (2012) 165–180
Contents lists available at SciVerse ScienceDirect
Developmental Review
journal homepage:
275). Although authorities assumed only children could provide details of crimes, adults had a critical
role in guiding the process (Trachtenberg, 1939). As one scholar explained, ‘‘the child would describe
in great detail what he saw in the oil—a demon dressed in a certain manner having a certain identifying
mark. Often the sorcerer would instruct the child to send that demon back and ask another one
to come, until the right demon appeared’’ (Dan, 1980, p. 28, emphasis ours).
If we replace ‘‘demons’’ with ‘‘memory,’’ ancient divination techniques are eerily similar to the use
of interviewing props when questioning children about suspected abuse. In both cases, adults believe
children can describe crimes but that special procedures are likely necessary to obtain reports. Both
traditions involve highly scripted instructions that increase interviewers’ confidence in the validity
of the process, and authorities trust that children’s innocence makes them unlikely to invent false reports.
Most important, interviewers focus children’s attention on objects while delivering specific—
sometimes suggestive—prompts.
Of course, authorities no longer claim that children have special abilities to connect with supernatural
forces, and no investigators use the shiny objects and incantations that characterized past practice.
Nonetheless, the parallels between ancient divination techniques and some current practices
raises an interesting question: If the confluence of children, objects, and questions has unanticipated
consequences, are some professionals still, in a sense, divining testimony?
Uses of objects in investigative interviews
Play materials such as crayons and paper, dolls, model houses, and sand boxes have been the staples
of psychodynamically oriented play therapy with children since the turn of the 20th century. Later
their use was extended to help children understand medical procedures, to show areas of pain on
their body, and for teaching purposes (e.g., to explain self-care practices). The idea that objects help
bridge the gap between what children know and what they can (or are willing to) explain is based
on the assumption that props help concrete thinkers grasp the point of conversations, increase the
completeness of reports by allowing children to respond without verbalizing embarrassing information,
and provide effective retrieval cues (Russell, 2008). For all of these reasons, many professionals
added props to their toolbox for investigative interviews of children (e.g., Hewitt, 1999; Jones & McQuiston,
1988; Morgan, 1995; for a brief history of protocols, see Poole & Dickinson, in press).
Throughout the 1980s and 1990s, many interviewers, interviewing centers, and some jurisdictions
had locally-developed scripts for when and how to use props.1 But due to concerns about children’s
suggestibility, since the early 1990s there have been many attempts to collect normative data on how
sexually abused and nonabused children behave with props and to examine the consequences of using
objects to elicit event reports.
In this paper, we explore whether two commonly used props, anatomical dolls and body (human
figure) diagrams, do have ‘‘special powers’’ to help children add important information to their autobiographical
reports. For each prop, we consider practitioners’ and policy makers’ responses to this
evidence and the degree to which evidence is currently being used to craft best practice guidelines.
We conclude by discussing the types of developmental research that are necessary to advance the field
of evidence-based interviewing.
Anatomically detailed (AD) dolls were created in the mid 1970s by three members of an interagency
rape team project (Morgan, 1995). The props quickly gained popularity and were one of the
major tools used to evaluate sexual abuse in preschool children, as reflected in a series of prominent
day care cases (e.g., the McMartin preschool trial, State of New Jersey v. Margaret Kelly Michaels; e.g.,
1 Today, the dozens of interviewing protocols guiding practice can be organized along a dimension anchored by opposing
philosophies regarding the value of props: One instructs interviewers to display props early in conversations to help elicit abuse
reports (Anderson et al., 2010), whereas the other recommends techniques for encouraging verbal reports without props (Lamb,
Hershkowitz, Orbach, & Esplin, 2008). In between lies typical practice, in which interviewers encourage verbal reports but
sometimes introduce props to clarify children’s reports or to probe for additional disclosures.
166 D.A. Poole, M. Bruck / Developmental Review 32 (2012) 165–180
see Rosenthal, 1995). According to an early review (Ceci & Bruck, 1993, p. 425), transcripts of therapy
sessions documented a disturbing array of practices, including ‘‘naming the dolls after defendants,
berating the dolls for alleged abuse against the children, assuming the role of fantasy characters in doll
play, and creating a persistent atmosphere of accusation’’. There was little doubt that some evaluators
used AD dolls in ways that divined testimony.
Outside the widely publicized cases, interviewers were using dolls for a multitude of purposes that
were summarized by Boat and Everson (1996): as an icebreaker (i.e., to convey that it was acceptable
to talk about sexual knowledge and experiences), to assess children’s names for body parts, as a diagnostic
‘‘screen’’ (to observe whether children’s interactions with dolls elicited statements or concerning
behavior that could then be explored), and to encourage children to demonstrate what happened
after they had provided verbal reports of abuse.
Regardless of the intended function of the dolls, there was an undocumented concern that some
children strayed into exploration and play with dolls during interviews. A parallel concern was that
interviewers became more suggestive when dolls were available, often bypassing open-ended questioning
and relying more heavily on specific questions paired with the dolls. The research that followed
to explore these issues fell into four broad categories: comparing how abused and
nonabused children interacted with AD dolls, evaluating how children reported documented experiences
with and without dolls, analyzing reports obtained by standard and doll-assisted investigative
interviews, and observing the impact of dolls on interviewers’ behavior. We now provide an overview
of the findings of this research.
Research findings
Normative data on how abused and nonabused children interacted with dolls during free play addressed
the important issue of whether dolls could be used to screen children who were likely abused.
Children’s interactions with AD dolls could be diagnostically useful if (a) sexualized handling was
much more common among children who had experienced abuse and (b) procedures for screening
out unfounded concerns were sufficiently accurate for forensic purposes. Such information is critical
if play patterns thought to identify abused children, such as manipulating the dolls in sexually explicit
ways and showing reticence or avoidance when presented with dolls, also occurred in samples of nonabused
children (see Ceci & Bruck, 1995, for a review).
Early evidence suggested that the diagnostic value of dolls was limited because many abused children
do not interact sexually with dolls, but dolls provide ‘‘affordances’’ (i.e., object features permitting
certain behavior) that could generate inaccuracies among both abused and nonabused children.
For example, dolls afford fondling genitalia out of curiosity, putting fingers in holes (think of what children
do with stuffed olives), and using the dolls as aggressive action figures. Findings regarding how
often children displayed these behaviors were varied yet coherent when aggregated: Fingering genitalia
and inserting fingers into holes were so common that some authors did not even code such
behaviors as sexual (e.g., Jampole & Weber, 1987; also see data from Cohn, 1991), children with no
known history of abuse infrequently displayed explicit sexual activity (e.g., intercourse) in the presence
of adults (e.g., 4% of a combined sample of 550+ children, Everson & Boat, 1997), some demographic
groups were more prone to this behavior (e.g., 27% of low SES black males in Everson &
Boat, 1990), and rates of sexualized behavior increased across demographic groups when adults left
the room (Everson & Boat, 1990).
Some doll supporters countered that the unexpected sexual behavior observed in studies could be
traced to pornography or other sources of sexual information (Dawson, Geddie, & Wagner, 1996), but
the fact that abused children may also have these experiences complicates the task of screening out
children whose knowledge stems only from these influences. Although there are still some dissenters,
the idea that spontaneous behavior with dolls has sufficient sensitivity and specificity to serve a diagnostic
function was the first concerning practice to lose widespread support: As early as 1990, the
American Professional Society on the Abuse of Children advised that AD dolls ‘‘should not be considered
a diagnostic test’’ (p. 5).
The second line of research on AD dolls was the most highly controlled: Children experienced
scripted touching in medical settings (e.g., genital examinations, annual check-ups, urological
D.A. Poole, M. Bruck / Developmental Review 32 (2012) 165–180 167
procedures) or laboratory events, and interviewers asked them to show what had happened with
dolls. Most important, their responses were sometimes compared to those of children interviewed
without the dolls or to children who were not touched. The resulting data could determine if dolls
helped children report actual physical contact without encouraging reports of target behavior that
had not occurred (e.g., genital touching or hitting), and also whether dolls added important accurate
information over and above what children related when merely asked questions without dolls.
Saywitz, Goodman, Nicholas, and Moan (1991) conducted one of the first studies using this paradigm.
Girls (ages 5 and 7 years) were questioned about a previous check-up in which some of the girls
had received a genital–anal examination whereas others had received a scoliosis examination. Direct
questions (‘‘Did that doctor touch you there?’’) paired with a doll greatly increased true reports of vaginal
touching without a similar increase in false reports (i.e., only 3% of children who had not experienced
vaginal touching falsely said that they had).
But results were not encouraging for younger boys and girls. In one pair of studies, 3-year-olds
(Bruck, Ceci, Francoeur, & Renick, 1995) and 4-year-olds (Bruck, Ceci, & Francoeur, 2000) received annual
medical check-ups by their pediatrician in which half the children experienced anal–genital exams.
Immediately after the physical exam, interviewers presented anatomical dolls along with
miniaturized versions of the nonmedical props, asked children to name the body parts, and then asked
specific questions about whether the doctor had touched various body parts. There were no age differences
in accuracy: Summing across age groups (see Bruck et al., 2000), 27% of the nontouched children
falsely claimed that the doctor had touched their genitalia/buttocks and 51% of the touched
children falsely denied being touched. When interviewers behaved more suggestively by asking children
to show on the doll how the doctor had touched their genitals and anal areas, 36% of the nontouched
children falsely showed touching. Even children who were touched made errors: 37% of all
their errors involved over-touching responses, such as inserting fingers into anal or genital cavities,
depicting experiences that had not occurred. These findings countered the prevailing wisdom that
it would be difficult to induce children to make false reports of genital touching.
There are several explanations for the discrepant findings of the Saywitz et al. and Bruck et al. studies.
The most obvious is age: As we discuss shortly, dolls pose greater risks to children younger than
5 years. Another possibility concerns an important procedural difference. Bruck et al. asked the children
to name body parts on the dolls, including the genitals and anal area, which is a common procedure
in prop-assisted interviews. Saywitz et al. did not follow this procedure—they merely showed
each child the doll, which was then undressed to reveal all the body parts. Because body part labeling
suggests (or at least primes) possible answers, this may be an important procedural difference that
explains why, in some contexts, props elicit a concerning rate of false reports.
Another study illustrates the need for nuanced discussions that acknowledge how the risks and
benefits of props shift as a function of children’s ages, their prior experiences, and interviewer practices.
Steward and Steward (1996) repeatedly interviewed children (3–6 years) about medical examinations
in which some had received genital and anal touches whereas others had not. Major analyses
compared children interviewed with verbal questions alone and those interviewed with props (combining
over doll and body diagram conditions).
This design, spanning ages and interviewing practices represented in the Bruck et al. and Saywitz
et al. studies, confirmed their collective results: Props increased accurate reports of embarrassing
touches among touched children—but at a cost. Specifically, the presence of props increased the percentage
of children who falsely reported forensically-meaningful touches, with the most errors occurring
when interviewers asked direct questions about touching to specific body parts. During a second
interview, for example, no child in the verbal condition falsely reported touching to the anal area, but
22% of the children interviewed with dolls did so when the prop accompanied initial questions, and
30% did so when interviewers asked ‘‘Were you touched there?’’ while pointing to the doll. (See Steward
& Steward, p. 95, for the discussion that separates doll and diagram conditions.) Surprisingly,
forensically-relevant commission errors sometimes involved school-aged children.
Before leaving this influential study, we caution readers that although these analog studies document
the benefits and risks of various interviewing techniques, they do not help policy makers
weigh these two types of results. This is because low rates of accurate disclosures in response to
questions alone were a predictable consequence of the fact that the children were not expecting
168 D.A. Poole, M. Bruck / Developmental Review 32 (2012) 165–180
discussions about touching and did not participate in interviews stacked with techniques to encourage
verbal reports (e.g., see pp. 163–166 in Steward and Steward, 1996). At the same time, the untouched
children were not in an atmosphere of concern about touching that would increase false
report rates. Together, these design features maximized the extent to which cuing with dolls (or
props in general) would increase true reports while providing lower-bound estimates on the frequency
of false reports. Results, therefore, do not speak to the ratio of benefits to risks when dolls
are added to front-line interviews, especially when the control condition (no dolls) represents current
best-practice standards.
The third type of study remedies these limitations by situating dolls in an investigative context. For
instance, Thierry, Lamb, Orbach, and Pipe (2005) examined forensic interviews of 3–12-year-old alleged
victims of sexual abuse who were provided with dolls at different parts of the interview. In response
to open-ended invitations, children interviewed with and without dolls reported equal
amounts of information, which bolsters confidence that interviewers had not introduced dolls only
for reticent children. Results showed that dolls served the desired memory function for older children
(7–12-year-olds), often triggering them to verbally report more information and to narrate while demonstrating.
The 3–6-year-olds, however, were more likely only to enact events. Thierry et al. expressed
concern about this ‘‘language-substitution’’ function because prior research found enactments to be
less accurate than the verbal reports that accompany dolls (Salmon, Bidrose, & Pipe, 1995). Also,
the young children sometimes played suggestively and provided more inconsistent details with dolls
than the older children did, and both age groups produced more fantastic details with dolls. The team
concluded that older children are better than younger children at reproducing reported information
with dolls and that, overall, doll use did not improve the amount or quality of information provided
by young children (see also Lamb, Hershkowitz, Sternberg, Boat, & Everson, 1996; Lindberg, Chapman,
Samsock, Thomas, & Lindberg, 2003; Pipe & Salmon, 2009).
A laboratory paradigm, which provided more control over how interviewers used dolls, found
converging evidence on the limited advantages of dolls. Salmon, Pipe, Malloy, and MacKay (2012,
Experiment 2) questioned 5–7-year-old children 1 week after they experienced an event involving
bodily contact. Whenever children reported having been touched, interviewers asked them to elaborate
either with a doll or using only the verbal prompts. There was no increase in errors associated
with dolls, but dolls did not elicit more information than questions alone. (For comparable findings,
see verbal response results in Goodman & Aman, 1990, and elaborative detail results in Gordon
et al., 1993.)
The final category of research centered on how interviewers use dolls. Two findings stood out in
one analysis of concerning practices (Boat & Everson, 1996). First, interviewers who introduced dolls
as a demonstration aid often did so prematurely, bringing dolls out for children who were not confused
or reticent, or saying such things as, ‘‘You don’t need to tell me—just show me’’ (p. 99; see also
Thierry et al., 2005). In other words, the availability of dolls led interviewers to terminate open-ended
questioning sooner. This point is important because doll advocates sometimes cite studies showing
that children provide additional details after introduction of dolls, but this finding is not informative
if the studies did not address what children would have reported had interviewers continued a line of
developmentally-appropriate questioning (i.e., when there was no control group involving an alternative
practice). Relatedly, another study found that interviewers spoke more when they were relying on
dolls compared to other portions of the interview, indicating that dolls can shift control of interviews
from children to adults (Santtila, Korkman, & Sandnabba, 2004; see Dickinson, Poole, & Bruck, 2005,
for an overview of concerns about AD dolls).
Another problem revealed in Boat and Everson’s (1996) analysis involved interviewers who used
dolls as a diagnostic screen. Not infrequently (e.g., 17% of interviews in which dolls were used as a
diagnostic screen with 2–5-year-olds), interviewers overinterpreted children’s behavior without asking
clarification questions (e.g., ‘‘Oh, that’s what happened to you’’; p. 100). In light of the problems
with dolls, many investigators have concluded that research and clinical efforts should focus on ways
to encourage verbal reports that do not involve object cues, such as building sufficient rapport and
including a warm-up phase before the actual interview to encourage children to talk (e.g., Thierry
et al., 2005).
D.A. Poole, M. Bruck / Developmental Review 32 (2012) 165–180 169
The mismatch between AD dolls, interviewers’ intent, and children’s minds
Given that children and interviewers seem to enjoy dolls, why have findings on the benefits of dolls
been disappointing? First, the results of basic developmental research, conducted years after the dolls
were introduced into forensic interviews, cast doubt on the feasibility of using dolls with very young
children. A prerequisite for conveying information about oneself with a doll is representational insight,
which is awareness that a doll is simultaneously an object and a symbol of a particular person.
Because young children find it difficult to grasp symbol-referent relations (DeLoache, 1991; DeLoache,
2000; DeLoache, 2005), Deloache and her colleagues thought they might also have difficulty using
dolls as symbols of themselves and others. In one these studies (Deloache & Marzolf, 1995), assistants
put stickers on children during a storybook activity and asked each child to show on a doll where the
stickers were placed. The youngest children (averaging 2.5 and 3.5 years) made many errors, even
though they were still wearing large stickers (59% and 29% of the placements were incorrect, respectively);
by 4 years of age, children made fewer errors (8% of the placements were incorrect).
But the fact that dolls are cognitively challenging for most children younger than 4 years does not
imply that older children always have the cognitive fundamentals to reenact events with dolls. The
ages-of-acquisition mentioned in research articles and reviews refer to the typical performance of study
volunteers, but there are older children with some impairment. In DeLoache and Marzolf’s (1995) study,
for example, 12.5% of 4-year-olds failed to accurately place at least three out of four stickers on a doll.
Clearly, this error rate is too high for forensic purposes. Performance on a second task, in which children
were touched during a Simon Says game, was more concerning: When interviewers asked children who
had just verbally reported an experienced touch to demonstrate on a doll, the 4-year-olds erred 19% of
the time. (Error rates for 2.5- and 3.5-year-olds who had just accurately described a touch were 47% and
33%, respectively). In another study, Deloache, Anderson, and Smith (cited in Uttal et al., 1998) asked
preschool teachers to record whenever someone did something to a child that was emotionally upsetting.
When the children were asked about these incidents either with or without dolls (using a fixed set
of questions), dolls produced no increase in the amount of information reported.
A largely silent issue is the special challenge of interviewing children with disabilities, which is a
group comprising about 16% of the victims involved in child protective services investigations (US
Department of Health, Administration for Children, Administration on Children, & Families, 2011).
One study of adults with mild and moderate mental retardation (Valenti-Hein, 2002) illustrates
why norms based on the typical children who participate in university research projects may lead policy
makers to overestimate the developmental appropriateness of props. When these adults placed
stickers on themselves in the same places that stickers were visible on a model, they were more accurate
when the target stickers appeared on a live model than on a doll or body diagram. For example,
the adults with moderate cognitive impairment accurately matched sticker placements from a live
model to themselves almost 80% of the time, but they were successful less than 65% of the time when
mapping between a doll and themselves. (See Deloache & Smith, 1999, for similar findings with
There is another reason why basic developmental research cannot set a fixed age when dolls are
safe for investigative interviews: Although representational insight is a necessary skill for competent
use of dolls, an array of other factors is involved in translating memories into accurate reports (Poole,
Bruck, & Pipe, 2011). As described earlier, interviewers sometimes combine props with questions that
elicit high rates of inaccurate information in laboratory studies, props make it easy to respond
thoughtlessly (e.g., by pointing or touching), and props are distracters that can lead children to lose
attention to the topic of questioning (e.g. Bruck et al., 2000; Thierry et al., 2005). As a result, there
is no basis for the claims that ‘‘dolls can be used with any child who can make the representational
shift’’ and are safe for most children ‘‘by the age of 5’’ (American Prosecutor’s Research Institute,
2004, p. 40).
Current practices and concerns
While the use of dolls was declining amongst growing criticism, some policy groups kept the door
open for limited use of dolls, creating ambiguities that allowed doll-assisted interviewing to thrive. In
170 D.A. Poole, M. Bruck / Developmental Review 32 (2012) 165–180
1991, a committee of the American Psychological Association acknowledged that ‘‘neither the dolls
nor their use are standardized or accompanied by normative data’’ yet concluded that dolls could
be used (Fox, 1991, p. 722). A working group later cautioned, however, that particular care should
be taken when interpreting results from children younger than 5 years (Koocher et al., 1995). Similarly,
the American Professional Society on the Abuse of Children (2002) concluded that AD dolls
‘‘can be an excellent communication/clarification tool, if used appropriately’’ (p. 6).
Today, the RATAC procedure (which is permitted for use in child advocacy centers in the US) allows
introduction of dolls in the interview after a report of abuse for clarification purposes, to demonstrate
consistency, to help children convey how touching occurred without using their own
bodies (i.e., distancing), and to assist them in communicating information they cannot or do not want
to verbalize (Anderson et al., 2010). These uses are authorized with the caveat that interviewers use
dolls only with children who can make the representational shift (as indicated by such things as successfully
choosing the doll that looks most like themselves and demonstrating on the doll rather than
themselves) and that interviewers avoid leading or suggestive questioning with dolls (Anderson et al.,
2010; Holmes, 2000). The continuing popularity of dolls despite the scientific findings cited above is
illustrated by a recent case review in which 49% of RATAC interviews used dolls (Hlavka, Olinger, &
Lashley, 2010).
On the surface, it may seem innocuous to introduce dolls after children have already reported inappropriate
touching. But consider how the assumptions behind this practice pertain to two groups of
children: one who was touched inappropriately and one who was not. For children who were touched,
accuracy rests on their ability to keep the immediate topic in mind throughout the doll episode (e.g.,
touching by a particular individual rather than other touches they have received), their skill in reproducing
received touches onto the doll without straying into overtouching responses, their ability to
refrain from irrelevant responses (intrusions) due to affordances provided by the doll, and the extent
to which responses mean the same to the child and observers.
In terms of the last requirement, this means that observers must accurately categorize doll-directed
behavior. This is not an easy task: In one study, two coders of children’s spontaneous behavior
with dolls disagreed 22% of the time about whether or not children were demonstrating sexual interactions
(Dawson et al., 1996). The difficulty of making such decisions is also consistent with the experience
of those who transcribe videotapes of interviews with children: Even with a full camera view, it
is sometimes very difficult to determine the actions performed with dolls. Consequently, even for
abused children, clarification with the dolls may not always remedy the ambiguities and misunderstandings
that can occur when adults talk to children.
What about children who were not touched? To understand why developmental researchers want
proof that dolls improve decision-making compared to other well-researched techniques, we first
need to explain how allegations are sometimes elicited. In one style of prop-assisted questioning,
interviewers ask children to label parts on a body diagram (including sexual body parts) prior to asking
a specific question about whether anyone has touched them in any places on the diagram. When
children respond ‘‘yes’’ (which is the only accurate answer, since all children have frequently been
touched in those places), interviewers ask them to point to where they were touched prior to another
specific question (such as ‘‘Who touched you there?’’). In essence, then, this is a ‘‘suggest-prime-point’’
procedure, one in which interviewers suggest sexual themes early in the interview (by displaying a
naked child), prime possible answers (by asking the child to label body parts that include desired answers),
open with a yes–no question (which young children answer unreliably in many contexts), and
then invite the child to respond by pointing. When interviewers bring dolls out for clarification, one
can assume they do so more frequently when children fail to provide coherent verbal narratives. These
circumstances should be common among those who were not abused but merely acquiesced to questions
due to thoughtlessness or exposure to prior conversations about abuse concerns.
The cognitive task for nonabused children does not involve mapping abusive events onto the dolls,
so reversing the course of the interview after dolls are brought out requires children to inhibit the urge
to respond due to a general tendency to take turns in conversations—a turn they may take, as some
children have done in laboratory studies, by demonstrating events that did not occur. The task is especially
challenging for children who must distinguish between memories of experienced events and
memories of prior conversations with concerned adults—that is, when memory source monitoring
D.A. Poole, M. Bruck / Developmental Review 32 (2012) 165–180 171
is key. From a cognitive perspective, it seems unlikely that adding objects that cue nonexperienced
events would facilitate source monitoring for age groups that frequently fail to spontaneously monitor
source (see Poole & Lindsay, 2001; Poole & Lindsay, 2002; Schacter, Kagan, & Leichtman, 1995).
Body diagrams
When prosecutors began facing challenges to doll-assisted interviews in court, some interviewers
reacted by substituting body diagrams. Diagrams come in different forms but typically are simple line
drawings of male and female children and adults. Some are clothed, some are naked, some graphically
depict male and female body parts, and some are gender neutral. Today, interviewers use body diagrams
for a variety of purposes, including learning the names children use for body parts, facilitating
discussions of good and bad touches, eliciting initial disclosures (when paired with questions such as,
‘‘Has anyone touched you in any of these places?), and to help children provide more or clearer information
about abuse (e.g., ‘‘Show me where you were touched.’’). Thus diagrams sometimes serve a
central role, as when interviewers display them early in interviews to elicit abuse disclosures (e.g.,
the RATAC protocol; Anderson et al., 2010), whereas other times interviewers display them later
in interviews to help clarify reports (and, sometimes, to probe for additional disclosures).
Interest in body diagrams increased after an interviewer training initiative in the U.S. (Finding
Words; Walters, Holmes, Bauer, & Vieth, 2003) selected a prop-focused procedure, the RATAC protocol
(aka the CornerHouse protocol), for widespread distribution.2 Despite the fact that RATAC was
developed in the late 1980s, until recently there was little scientific research on the benefits, risks,
and incremental validity of diagram-assisted questioning. Nonetheless, diagrams were thought to be
safer than dolls because two-dimensional props lack orifices that can be penetrated out of curiosity. Also,
some argued that children find it easier to use drawings to represent themselves than dolls. The idea was
that children have extensive experience with pictorial representations, but also that line drawings are
less salient as objects, which could improve children’s ability to use them as symbols (Russell, 2008).
In addition to the assumption that diagrams are more developmentally appropriate than dolls, another
reason for their widespread popularity is that RATAC promoters claimed their protocol was evidencebased
(Anderson et al., 2010). We discuss children’s reactions to body diagrams and concerns first, and
then turn to the required criteria for claiming that a protocol is evidence-based.
Children’s performance with body diagrams
Is it true that the representational task is easier with body diagrams than dolls, resulting in more
accurate demonstrations of touching? Consider data from an in-progress study (Ladd, London, &
Bruck, 2011). A research assistant asked each child to identify body parts on a clothed body diagram,
a doll, and themselves (e.g., ‘‘The doll has a hand. Can you point to the doll’s hand?’’). Next, she placed
a large sticker on one of the child’s body parts and asked the child to place a similar but smaller sticker
on the doll, the body diagram, or the assistant. Each sticker remained on the child throughout the response
and was then removed.
Results confirmed previous findings that 3-year-olds are flummoxed by such tasks: This group was
correct only 66% of the time with dolls and 45% of the time with body diagrams. Five-year-olds performed
well, making accurate placements 99% of the time on dolls and 95% of the time on diagrams. As
expected, 4-year-olds were a transitional group; their accuracy rate was 84% in the doll condition and
79% in the body diagram condition. Several points are forensically relevant. First, the stickers were still
visible, so this task may be easier than mapping information that resides only in memory (e.g., compare
sticker task and Simon Says results from DeLoache & Marzolf, 1995). Therefore, the relative success
of the 5-year-olds should not be taken to mean that props are appropriate when this age group
performs a pure memory task. Second, performance was worse with body diagrams, not better. In the
aforementioned study of adults with cognitive impairments (Valenti-Hein, 2002), there was also no
consistent benefit of diagrams over dolls. Thus this study does not support the claim that children
are more successful at mapping body touches on line drawings than dolls.
2 This initiative continues today as ChildFirst™ (National Child Protection Training Center, 2012).
172 D.A. Poole, M. Bruck / Developmental Review 32 (2012) 165–180
A pair of studies by Brown and colleagues (Brown, Pipe, Lewis, Lamb, & Orbach, 2007; Brown, Pipe,
Lewis, Lamb, & Orbach, 2011) confirmed that even 5–7-year-olds find it cognitive challenging to report
events on body diagrams. To help the children understand the purpose of the body diagram, interviewers
delivered practice instructions: They touched children on the elbow and then asked them
to show on the diagram where they were touched. Although no child was under 5 years of age in this
study, 54% of the children required correction and additional explanation to successfully complete a
These alarming results raise an obvious question: Why did body diagrams become popular before
basic safety testing had been conducted? Perhaps because diagrams elicit reports of touching, which is
very rewarding to interviewers. Consider two studies that analyzed investigative interviews of alleged
abuse victims conducted with the NICHD protocol, a procedure that encourages verbal reports through
rapport-building, a practice interview, and open-ended questions that invite children to do most of the
talking (Aldridge et al., 2004; Teoh, Yang, Lamb, & Larsson, 2010). After interviewers believed they
could elicit no more information using the protocol, they asked questions referring to unclothed, gender-neutral
diagrams, including specific yes–no questions (e.g., ‘‘Did s/he touch this part?’’) followed
by open-ended prompts that invited children to elaborate. In both studies, adding diagram-assisted
questioning led children to recall additional information, including elaborations of bodily contact,
and this was particularly the case for younger children.
Two caveats are important in interpreting these results. First, the accuracy of information is usually
unknown in field studies, leading to the possibility that the added information produced by body diagrams
and specific questions included false information that decreased the overall accuracy of reports.
In fact, this is exactly what a laboratory study that mimicked the design of the field studies found (Otgaar,
Horselenberg, van Kampen, & Lalleman, 2011). Second, children and adults alike tend to provide
more information with additional recall attempts, so these studies do not reveal what children would
have reported had interviewers shifted gears in some other way (for example, by taking a break or
turning briefly to neutral topics) before asking additional questions. The value of these studies is to
document that children often provide more information when questioning continues, even after an
extensive interview, but the results do not speak to the forensic value of diagram-assisted questioning.
Recognizing the inherent limitations of field studies (specifically the inability to establish children’s
accuracy), researchers turned to more controlled situations in which children were touched during
scripted events and then asked to recall where they had been touched either with or without body
diagrams. In a set of studies we mentioned earlier, Brown and colleagues studied 5–7-year-old children
who were touched seven times by a photographer and then were asked to recall 1 month (Brown
et al., 2007) and 7 months (Brown et al., 2011) after the event. The children were questioned
extensively using a structured verbal protocol and then randomized to one of three conditions: verbal
questions only (an initial question about touching followed by open-ended inquiry and then specific
yes–no questions), verbal questions with body diagrams, and verbal questions with diagrams
preceded by practice instructions. At both the 1 month and 7 month interviews, children recalled very
little information about touching, regardless of condition. In the 1 month interview, the major finding
was that the addition of the diagrams did not promote additional accurate details beyond what
children in the nondiagram condition achieved; however, the children made more errors in the body
diagram conditions. At the 7-month follow-up, children remained in their original groups and were
questioned in the same ways as before. Accuracy rates were strikingly different for children who produced
any correct information during both interviews: 70% of the information elicited in the verbal
questions only condition was accurate at the 7-month interview, compared to only 23% in the diagram
with instruction condition and 22% in the diagram only condition.
Bruck (2009) conducted similar studies but included younger children (3–7 years) and manipulated
the placement of the body diagram inquiry viz a viz the verbal inquiry: Half the children received
a verbal interview followed by a body diagram interview containing the same questions as in Brown
et al., and half the children received the body diagram interview first, followed by the verbal-only
interview. Given the poor levels of recall in the Brown et al. study at 1 month, children were questioned
1 week after a touching activity with a magician for Study 1, and immediately after that
scripted activity for Study 2. In both studies and in both interviewing conditions, children recalled
few touches in response to prompts about touching, and diagrams did not increase accurate recall.
D.A. Poole, M. Bruck / Developmental Review 32 (2012) 165–180 173
However, diagrams did increase error rates when interviewers presented them after the verbal
Clearly, these studies did not find incremental validity from adding diagrams to the interview.
Additional evidence is provided by Salmon et al. (2012), who also found that adding diagrams to questions
did not elicit additional elaboration or clarification of details. Given these findings, the potential
for false reports provided by diagrams is even more concerning. That is, just as anatomical dolls permit
children to inspect genitals and insert fingers in holes, body diagrams make it easy for children to
point to body parts that were not touched during target experiences. This is true not only for very
young children but, in the study we describe next, for school-aged children as well.
To embed diagrams in realistic investigative protocols, Poole and Dickinson (2011) watched actual
prop-assisted interviews and then modeled the instructions to create a body-diagram-focused procedure
(i.e., one in which interviewers used diagrams to elicit initial reports of touching). The comparison
condition in this study (the standard condition) included the same warm-up questions used in
the diagram-assisted procedure (only delivered with no interviewing aids), followed by open-ended
and then more focused questions (still with no specific mention of touching). Half the children (age
4–9 years) had experienced touching (which involved mishaps to make the touches more memorable);
the other half only heard about these events through a story their parents read to them that described
nonexperienced touching. Therefore, this study compared the number of true and false reports
of touching across two procedures, including a popular way of using diagrams in prop-assisted forensic
Poole and Dickinson (2011) found that no child disclosed experienced touching during open-ended
questioning in the standard condition, whereas pairing body diagrams with specific questions about
touching did elicit disclosures in a minority of children.3 However, there was no significant advantage
of body diagram interviews after interviewers in the standard condition progressed to more focused
questions, such as ‘‘Did Mr. Science give you anything in the science room?’’ and ‘‘Tell me all about
(something the child mentioned). What happened with/when (the child’s words)?’’ In addition, body diagram
questioning elicited more false reports of suggested touches than did open-ended questioning in
the standard condition, and only body diagram questioning prompted intrusions (spontaneous false reports
that were not suggested, which was a type of error found across all ages). Thus, embedding diagrams
in a research protocol similar to those used in the field returned the same results discussed
earlier: Specific questions with diagrams yielded more true reports than open-ended questioning did,
but with an increase in the number of false reports. Importantly, this study underscores two ways in
which diagram-assisted questioning about touching increases errors: by permitting thoughtless pointing
(i.e., spontaneous intrusions) and by drawing out more suggested touches than open-ended questions
delivered without a diagram. Poole and Dickinson concluded that body diagrams paired with yes–no
questions are more risky than yes–no questions alone because ‘‘yes–no questions rarely prompt children
to describe events that are unrelated to information embedded in the questions’’ (p. 668).
Some practitioners are critical of the analog research described thus far because (a) children do not
regularly make forensically relevant errors about touching, (b) the studies did not include questions
about inappropriate genital or anal touching, and (c) interviewers did not question children who
had already disclosed (which is common in forensic settings). Perhaps under these circumstances,
risk-benefit ratios associated with diagrams would be more favorable.
In terms of the first two issues, it should be noted that although forensically-relevant errors (i.e.,
false reports of touch to the genitals, breasts, or anal area) are low in analog studies, they do occur
(for example, see Brown et al., 2007). Their infrequency does not speak to ability of diagrams to help
children report events, however, because the diagrams in many analog studies did not depict sexual
parts and the children were not in an atmosphere of concern about sexual touching.
3 There are a number of reasons why children who have not previously disclosed touching sometimes show low rates of
disclosure in response to direct questions about touching: Some children may be reluctant to talk about touching, the word
‘‘touch’’ may not be a good memory cue (Poole et al., 2011), questions with the word ‘‘any’’ (as in ‘‘Did anyone touch you in any
….’’) are notorious for returning ‘‘no’’ responses (Walker, 1999), and booting up one or two instances of touching into memory may
temporarily inhibit retrieval of other instances.
174 D.A. Poole, M. Bruck / Developmental Review 32 (2012) 165–180
In terms of the third issue, efforts to explore whether previous disclosures shift findings are still in
the infancy phase, but a procedure is under development. Recently, Dickinson and Poole (2011) created
a new paradigm, Mr. Science—Germ Detective! This procedure capitalizes on the fact that even
young children have strong emotional and behavioral reactions to contamination. For example, 2-
and 3-year-olds usually fail appearance-reality tests but do well when the hidden attribute is moldy
bread (Siegal & Share, 1990), and even 2-year-olds frequently refuse to touch contaminated objects
(Stevenson, Oaten, Case, Repacholi, & Wagland, 2010; see also Curtis & Biran, 2001). As a result, innocuous
laboratory touches can be converted into highly memorable touches by telling children these
touches are potentially contaminating.
In a pilot study, assistants instructed 26 4–8-year-olds that Mr. Science was not allowed to touch
children’s skin (because this is how germs spread), which provided opportunities for inappropriate,
memorable touching. When Mr. Science twice attempted to break ‘‘the germ rule’’ during a set of
fun activities about germs, more than a third of the children blocked a skin touch from occurring, more
than a quarter spontaneously reported touching to a parent, and almost half of the nondisclosers reported
a target touch when a researcher spoke to children on the phone prior to final interviews (Dickinson
& Poole, 2011). Clearly, these laboratory touches (or attempted touches) were salient.
All children participated in the same interviewing condition in which interviewers began with
open-ended questions and progressed through phases representing increasingly focused questions
about touching that included body diagrams. Although final results await completion of a multi-site
study, preliminary findings confirmed the feasibility of using this paradigm to investigate children’s
reports of memorable touches. First, children’s disclosure histories (i.e., whether children told a parent
or researcher prior to the interview or not) predicted disclosure during the formal interview: The
majority of children who previously disclosed inappropriate touching did so during the follow-up
interview, whereas only a minority of the nondisclosers did. This is an important finding because it
mimics patterns from case reviews (see Hershkowitz, Horowitz, & Lamb, 2005; London, Bruck, Ceci,
& Shuman, 2007).
A second finding confirmed results from previous studies reported in this paper: Some children falsely
said ‘‘yes’’ to specific questions paired with diagrams, and reports of nonexperienced touches
were as frequent as new accurate reports when interviewers progressed to diagram-assisted questions.
What is striking is the consistent finding across paradigms that some young children have difficulty
resisting specific questions about touching paired with body diagrams (also see Steward and
Steward, 1996).
In sum, body diagrams are not developmentally more appropriate than dolls: Young children often
find it challenging to accurately reproduce events on diagrams, diagrams do not consistently increase
reports of experienced touches beyond what can be achieved with verbal questions alone, and diagram-assisted
questioning increases false reports. Although no studies have compared the consequences
of preceding questions about touching with body part labeling, the increase in errors
associated with body diagrams has occurred even when procedures did not include this preliminary
phase (e.g., Brown et al., 2007).
Beliefs maintaining diagram-focused procedures
Why have diagram-focused procedures survived for over 20 years while protocols based on memory
and interviewing research were available without cost and were widely distributed during this
same time period? (For examples of research on alternative approaches, see Holliday, 2003; Holliday
& Albon, 2004; Köhnken, Milne, Memon, & Bull, 1999; and Lamb et al., 2008.) We think the reasons
involve practical as well as ideological issues. On the practical side, it is difficult to train interviewers,
but diagram-focused procedures are straightforward and appealing to interviewers and trainers alike.
Also, these procedures have been used in thousands of adjudicated cases, which circuitously has been
used to justify their value. Supporters defend that value by saying that (a) young children have limited
attention spans and problems with memory recall that justify getting to the point early in interviews
by cuing touch reports (because this will increase sensitivity—the detection of true cases of abuse),
and (b) good investigative practices will screen out the few false reports elicited by cuing (so specificity
is not a problem).
D.A. Poole, M. Bruck / Developmental Review 32 (2012) 165–180 175
As intuitively appealing as these claims are, there is no evidence to support them. Young children
do have short attention spans, but this does not mean that raising sensitive issues early in conversations
with unfamiliar adults increases disclosures compared to alternative procedures. Regarding
specificity, there is evidence that research-based protocols improve credibility assessments (Hershkowitz
et al., 2007), but we know of no evidence on the accuracy of professional assessments of disclosures
elicited by prop-based cuing.
Beliefs notwithstanding, prop-assisted interviewing will only be elevated to evidence-based practice
when particular uses of objects have passed the tests society typically requires to market interventions.
Consider next how props fit into this process.
What is evidence in evidence-based interviewing?
The most striking feature of current debates over interviewing props involves the concept of ‘‘scientifically
based’’: Those who distribute prop-based protocols claim these approaches are grounded in
research (e.g., Anderson et al., 2010), whereas critics claim they are not (Poole et al., 2011). As confusing
as this discrepancy may be to policy makers and the courts, the issues involved have already been
hashed out in medicine and clinical psychology, where similar assessment and intervention goals
guide practice. Briefly, the gold standard of evidence is reviews (including meta-analyses) of randomized
controlled trials (e.g., Cochrane Database of Systematic Reviews) that consider the statistical and
clinical significance of interventions (among other issues; e.g., generalizability across individuals
and environments with various characteristics). However, the ‘‘best research evidence’’ at one point
in time may lie anywhere along a continuum of evidence (American Psychological Association,
2005; Stuart & Lilienfeld, 2007).
The structure of this continuum can be illustrated by considering the phases of testing that occur
before drugs come to market. After sufficient preclinical research (basic research) justifies the approach
promised by a drug, Phase I (human microdosing) studies begin to assess the drug’s safety
and efficacy through the administration of limited dosages to small numbers of healthy volunteers.
This is followed by subsequent testing with larger samples of non-patient volunteers and patients
(Phases II through IV; US Department of Health & National Institutes of Health, 2007). Unlike earlier
phases, later phases compare the drug to existing treatments and evaluate the drug’s performance in
the environments in which it will be used. In intervention research more broadly, it is well understood
that lower-phase studies are necessary steps that should not be bypassed. In other words, even if
researchers could jump directly to a randomized-control field trial, it would be imprudent to do so because
this research provides inadequate information about how and why results are occurring (Rogers,
To protect the public, nonmedical interventions should also require a systematic approach to safety
and efficacy testing. The danger of marketing interventions without sufficient evidence is illustrated
by the numerous psychological therapies that inadvertently caused harm, and this can occur even
when the children and adults who receive interventions appear to be helped. For example, without
adequate control conditions we do not know if people would have fared as well or even better had
they received a different intervention or no intervention at all (see Lilienfeld, 2007, for a selective
How do the phases of intervention research apply to the field of investigative interviewing? First,
basic research involves information about whether a technique is within the competence of children
in specific age ranges. These foundational safety tests include studies of whether children have the
cognitive ability to represent events with objects (considering, as part of the package, their ability
to understand the questions accompanying props). Because many factors influence performance
and shift results from study to study, extrapolations from basic research are best made to help interviewers
choose the most efficacious and least risky procedure from a pool of possible procedures. Thus
interviewing protocols that were built from the ground up (that is, by starting with basic research on
language and cognitive development and then considering studies of eyewitness testimony) are evidence
based because their originators looked at a variety of findings and chose techniques associated
with the highest success rates across all ages (Poole & Dickinson, in press).
176 D.A. Poole, M. Bruck / Developmental Review 32 (2012) 165–180
Regarding props, the basic research reviewed here shows that objects should not be used as symbols
with 3-year-olds, that error rates among 4-year-olds are high for forensic purposes, and that even
older children sometimes have difficulty with basic representational tasks. In the future, these types of
studies should be reproduced with groups of children who more closely resemble those involved in
maltreatment investigations, and data reporting should include the percentage of children who are
accurate across several trials.
The analog studies we reviewed represent the next stage of research (Phase I). Here props are
tested on nonabused (non-patient) samples to begin to identify benefits and risks in tasks containing
more challenges of memory testing (which is not possible with alleged victims because actual experiences
are unknown). One unexpected finding from these studies is the lack of consistent evidence
that props increase disclosures beyond what can be obtained from verbal questions alone. In-progress
research is trying to identify situations where props might be justified through paradigms that
better mimic forensic realities and by embedding props in interviews resembling front-line
If results emerge to support some uses of props, the next step should be controlled trials with alleged
victims. Because the instruments are to be used for diagnostic purposes, these trials would explore
the sensitivity and specificity of the major approaches by including cases where there is strong
evidence of abuse as well as cases deemed unlikely to be true. After comparing protocols to determine
what is ‘‘best in town,’’ the final studies would also observe how well interviewers use the techniques
and retain fidelity over time. Ultimately, only controlled comparisons of techniques with and without
props will move the field beyond long-held assumptions and into evidence-based practice.
Concluding comments and a call for innovation
Giving children a voice means interviewing them in ways that are developmentally sensitive: This
includes front-loading interviews with techniques that encourage sensitive disclosures, using words
and questions they understand, and realizing how easily children drift off topic so precautions are followed
to avoid misunderstandings. Before adding props to this difficult process, the field of maltreatment
investigation needs to establish that responses with props are at least as good a reflection of
event memories as responses to verbal questions alone—both for children who have experienced
abuse and for those who have not.
The idea of presenting ‘‘developmentally sensitive’’ or ‘‘engaging’’ techniques to draw children into
an interview does, on the surface, seem promising. After more that two decades of research, however,
we are learning that children have different ideas than adults about what that entails. To date, the
available evidence does not instill confidence that props are the best solution to the challenges of
interviewing children. Recall that immediately after interviewers touched children’s elbows, the
majority of 5–7-year-olds could not accurately touch a body diagram to reproduce this event (Brown
et al., 2007; Brown et al., 2011), young children who experienced genital touching sometimes strayed
into overtouching responses when handling AD dolls (Bruck et al., 1995; Bruck et al., 2000), and there
is no clear evidence that props provide benefits that cannot be gained from verbal questioning alone.
Although props are especially worrisome when used with preschool children, even school-aged children
sometimes respond off topic when interviewers add props to direct questions (Poole & Dickinson,
The research presented here does not imply that dolls and diagrams are intrinsically evil. Rather,
as with scrying, some characteristics of children, when combined with practices that divert attention
from internal cognitions to external objects and adult speech, shift the focus of control away
from children and onto objects and adults. A challenge for the future is to understand how each
of these components—mere attention to objects, the features of objects, and the instructions from
adults—promote accurate and inaccurate disclosures for children with different event and recall histories.
As some researchers pursue these questions, we hope others will investigate novel approaches
for helping children explain their experiences—approaches that move beyond long held
assumptions so we can improve our development and delivery of developmentally sensitive
D.A. Poole, M. Bruck / Developmental Review 32 (2012) 165–180 177
Preparation of this article, and some research summarized in it, was supported by the National Science
Foundation, under Grant Nos. SES-0718889 and SES-1121873 (DP), and the Institute of Child
Health and Human Development under Grant No. 5R01HD52034 (MB).
Aldridge, J., Lamb, M. E., Sternberg, K. J., Orbach, Y., Esplin, P. W., & Bowler, L. (2004). Using a human figure drawing to elicit
information from alleged victims of child sexual abuse. Journal of Consulting and Clinical Psychology, 72, 304–316.
American Professional Society on the Abuse of Children (1990). Guidelines for psychosocial evaluation of suspected sexual abuse in
young children. Author.
American Professional Society on the Abuse of Children (2002). Practice guidelines: Investigative interviewing in cases of alleged
child abuse. Author.
American Prosecutors Research Institute, National Center for the Prosecution of Child Abuse (2004). Investigation and
prosecution of child abuse. Thousand Oaks, CA: Sage.
American Psychological Association (2005). Policy statement of evidence-based practice in psychology. <
Anderson, J., Ellefson, J., Lashley, J., Miller, A. L., Olinger, S., Russell, A., et al (2010). The CornerHouse forensic interview protocol:
RATAC. Thomas M. Cooley Journal of Practical and Clinical Law, 12, 193–331.
Beard, M., North, J. A., & Price, S. R. F. (1998). Religions of Rome: A history. Cambridge: Cambridge University Press.
Bilu, Y. (1981). Pondering ‘‘The Princes of the Oil’’: New light on an old phenomenon. Journal of Anthropological Research, 37,
Boat, B. W., & Everson, M. D. (1996). Concerning practices of interviewers when using anatomical dolls in child protective
services investigations. Child Maltreatment, 1, 96–104.
Brown, D. A., Pipe, M.-E., Lewis, C., Lamb, M. E., & Orbach, Y. (2011). How do body diagrams affect the accuracy and consistency
of children’s reports of bodily touch across repeated interviews? Applied Cognitive Psychology.
Brown, D. A., Pipe, M.-E., Lewis, C., Lamb, M. E., & Orbach, Y. (2007). Supportive or suggestive: Do human figure drawings help 5–
7-year-old children to report touch? Journal of Consulting and Clinical Psychology, 75, 33–42.
Bruck, M. (2009). Human figure drawings and children’s recall of touching. Journal of Experimental Psychology: Applied, 15,
Bruck, M., Ceci, S. J., & Francoeur, E. (2000). Children’s use of anatomically detailed dolls to report genital touching in a medical
examination: Developmental and gender comparisons. Journal of Experimental Psychology: Applied, 6, 74–83.
Bruck, M., Ceci, S. J., Francoeur, E., & Renick, A. (1995). Anatomically detailed dolls do not facilitate preschoolers’ reports of a
pediatric examination involving genital touching. Journal of Experimental Psychology: Applied, 1, 95–109.
Ceci, S. J., & Bruck, M. (1993). Suggestibility of the child witness: A historical review and synthesis. Psychological Bulletin, 113,
Ceci, S. J., & Bruck, M. (1995). Jeopardy in the courtroom: A scientific analysis of children’s testimony. Washington, DC: American
Psychological Association.
Cohn, D. S. (1991). Anatomical doll play of preschoolers referred for sexual abuse and those not referred. Child Abuse & Neglect,
15, 455–466.
Curtis, V., & Biran, A. (2001). Dirt, disgust, and disease: Is hygiene in our genes? Perspectives in Biology and Medicine, 44, 17–31.
Dan, J. (1980). Samael, Lilith, and the concept of evil in early Kabbalah. Association for Jewish Studies Review, 5, 17–40.
Dawson, B., Geddie, L., & Wagner, W. (1996). Low-income preschoolers’ behavior with anatomically detailed dolls. Journal of
Family Violence, 11, 363–378.
DeLoache, J. S. (1991). Symbolic functioning in very young children: Understanding of pictures and models. Child Development,
62, 736–752.
DeLoache, J. S. (2000). Dual representation and young children’s use of scale models. Child Development, 71, 329–338.
DeLoache, J. S. (2005). Mindful of symbols. Scientific American, 293, 72–77.
DeLoache, J. S., & Marzolf, D. P. (1995). The use of dolls to interview young children: Issues of symbolic representation. Journal of
Experimental Child Psychology, 60, 155–173.
DeLoache, J. S., & Smith, C. M. (1999). Early symbolic representation. In I. E. Sigel & I. E. Sigel (Eds.), Development of mental
representation: Theories and applications (pp. 61–86). Mahwah, NJ: Lawrence Erlbaum.
Dickinson, J. J., & Poole, D. A. (2011). Mr. Science—Germ Detective! A novel paradigm for manipulating disclosure histories for
research on children’s eyewitness testimony. In Paper presented at the 4th Conference of the International Investigative
Interviewing Research Group (iIIRG). Dundee, Scotland, June.
Dickinson, J. J., Poole, D. A., & Bruck, M. (2005). Back to the future: A comment on the use of anatomical dolls in forensic
interviews. Journal of Forensic Psychology Practice, 5, 63–74.
Everson, M. D., & Boat, B. W. (1990). Sexualized doll play among young children: Implications for the use of anatomical dolls in
sexual abuse evaluations. Journal of the American Academy of Child and Adolescent Psychiatry, 29, 736–742.
Everson, M. D., & Boat, B. W. (1997). Anatomical dolls in child sexual abuse assessments: A call for forensically relevant research.
Applied Cognitive Psychology, 11, S55–S74.
Fox, R. E. (1991). Proceedings of the American Psychological Association, incorporated, for the year 1990: Minutes of the annual
meeting of the Council of Representatives. American Psychologist, 46, 689–726.
Goodman, G. S., & Aman, C. (1990). Children’s use of anatomically detailed dolls to recount an event. Child Development, 61,
Gordon, B. N., Ornstein, P. A., Nida, R. E., Follmer, A., Crenshaw, M. C., & Albert, G. (1993). Does the use of dolls facilitate
children’s memory of visits to the doctor? Applied Cognitive Psychology, 7, 459–474.
178 D.A. Poole, M. Bruck / Developmental Review 32 (2012) 165–180
Hershkowitz, I., Fisher, S., Lamb, M. E., & Horowitz, D. (2007). Improving credibility assessment in child sexual abuse allegations:
The role of the NICHD investigative interview protocol. Child Abuse & Neglect, 31, 99–110.
Hershkowitz, I., Horowitz, D., & Lamb, M. E. (2005). Trends in children’s disclosure of abuse in Israel: A national study. Child
Abuse & Neglect, 29, 1203–1214.
Hewitt, S. K. (1999). Assessing allegations of sexual abuse in preschool children: Understanding small voices. Thousand Oaks, CA:
Hlavka, H. R., Olinger, S. D., & Lashley, J. L. (2010). The use of anatomical dolls as a demonstration aid in child sexual abuse
interviews: A study of forensic interviewers’ perceptions. Journal of Child Sexual Abuse, 19, 519–553.
Holliday, R. E. (2003). Reducing misinformation effects in children with cognitive interviews: Dissociating recollection and
familiarity. Child Development, 74, 728–751.
Holliday, R. E., & Albon, A. (2004). Minimising misinformation effects in young children with cognitive interview mnemonics.
Applied Cognitive Psychology, 18, 263–281.
Holmes, L. S. (2000). Using anatomical dolls in child sexual abuse forensic interviews. Update, 13 (8). <
Jampole, L., & Weber, M. K. (1987). As assessment of the behavior of sexually abused and nonsexually abused children with
anatomically correct dolls. Child Abuse & Neglect, 11, 187–192.
Jones, D. P. H., & McQuiston, M. G. (1988). Interviewing the sexually abused child. London: Gaskell.
Köhnken, G., Milne, R., Memon, A., & Bull, R. (1999). The cognitive interview: A meta-analysis. Psychology, Crime & Law, 5, 3–27.
Koocher, G. P., Goodman, G. S., White, C. S., Friedrich, W. N., Sivan, A. B., & Reynolds, C. R. (1995). Psychological science and the
use of anatomically detailed dolls in child sexual-abuse assessments. Psychological Bulletin, 118, 199–222.
Ladd, N. E., London, K., & Bruck, M. (2011). Children’s developing understanding of dolls and drawings as symbols to map body touch.
In Paper presented at the meeting of the American Psychology-Law Society, Miami, FL, March.
Lamb, M. E., Hershkowitz, I., Orbach, Y., & Esplin, P. W. (2008). Tell me what happened: Structured investigative interviews of child
victims and witnesses. West Sussex, England: Wiley.
Lamb, M. E., Hershkowitz, I., Sternberg, K. J., Boat, B., & Everson, M. D. (1996). Investigative interviews of alleged sexual abuse
victims with and without anatomical dolls. Child Abuse & Neglect, 20, 1251–1259.
Lilienfeld, S. O. (2007). Psychological treatments that cause harm. Perspectives on Psychological Science, 2, 53–70.
Lindberg, M. A., Chapman, M. T., Samsock, D., Thomas, S. W., & Lindberg, A. W. (2003). Comparisons of three different
investigative interview techniques with young children. Journal of Genetic Psychology, 164, 5–28.
London, K., Bruck, M., Ceci, S. J., & Shuman, D. W. (2007). Disclosure of child sexual abuse: A review of the contemporary
empirical literature. In M.-E. Pipe, M. E. Lamb, Y. Orbach, & A.-C. Cederborg (Eds.), Child sexual abuse: Disclosure, delay, and
denial (pp. 11–39). New York: Routledge.
Morgan, M. (with Edwards, V.) (1995). How to interview sexual abuse victims. Thousand Oaks, CA: Sage.
National Child Protection Training Center (2012). ChildFirst™ states & countries. <
Otgaar, H., Horselenberg, R., van Kampen, R., & Lalleman, K. (2011). Clothed and unclothed human figure drawings lead to more
correct and incorrect reports of touch in children. Psychology, Crime, & Law.
Pipe, M.-E., & Salmon, K. (2009). Dolls, drawings, body diagrams, and other props: Role of props in investigative interviews. In K.
Kuehnle & M. Connell (Eds.), The evaluation of child sexual abuse allegations: A comprehensive guide to assessment and
testimony (pp. 365–395). Hoboken, NJ: Wiley.
Poole, D. A., & Dickinson, J. J. (in press). Investigative interviews of children. In R. Holliday & T. Marche (Eds.), Child forensic
psychology. Houndmills, Basingstoke, Hampshire: Palgrave Macmillan.
Poole, D. A., Bruck, M., & Pipe, M.-E. (2011). Forensic interviewing aids: Do props help children answer questions about
touching? Current Directions in Psychological Science, 20, 11–15.
Poole, D. A., & Dickinson, J. J. (2011). Evidence supporting restrictions on uses of body diagrams in forensic interviews. Child
Abuse & Neglect, 35, 654–669.
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.
Poole, D. A., & Lindsay, D. S. (2002). Reducing child witnesses’ false reports of misinformation from parents. Journal of
Experimental Child Psychology, 81, 117–140.
Rogers, M. A. (2009). What are the phases of intervention research? Access Academics and Research. <
questions/PhasesClinicalResearch/>, June.
Rosenthal, R. (1995). State of New Jersey v. Margaret Kelly Michaels: An overview. Psychology, Public Policy, and Law, 1, 246–271.
Russell, A. (2008). Out of the woods: A case for using anatomical diagrams in forensic interviews. Update, 21 (1), 2–6. <http://>.
Salmon, K., Bidrose, S., & Pipe, M.-E. (1995). Providing props to facilitate children’s event reports: A comparison of toys and real
items. Journal of Experimental Child Psychology, 60, 174–194.
Salmon, K., Pipe, M.-E., Malloy, A., & MacKay, K. (2012). Do non-verbal aids increase the effectiveness of ‘best practice’ verbal
interview techniques? An experimental study. Applied Cognitive Psychology, 26, 370–380.
Santtila, P., Korkman, J., & Sandnabba, N. K. (2004). Effects of interview phase, repeated interviewing, presence of a support
person, and anatomically detailed dolls on child sexual abuse interviews. Psychology, Crime & Law, 10, 21–35.
Saywitz, K. J., Goodman, G. S., Nicholas, E., & Moan, S. F. (1991). Children’s memories of a physical examination involving genital
touch: Implications for reports of child sexual abuse. Journal of Consulting and Clinical Psychology, 59, 682–691.
Schacter, D. L., Kagan, J., & Leichtman, M. D. (1995). True and false memories in children and adults: A cognitive neuroscience
perspective. Psychology, Public Policy, and Law, 1, 411–428.
Siegal, M., & Share, D. L. (1990). Contamination sensitivity in young children. Developmental Psychology, 26, 455–458.
Stevenson, R. J., Oaten, M. J., Case, T. I., Repacholi, B. M., & Wagland, P. (2010). Children’s response to adult disgust elicitors:
Development and acquisition. Developmental Psychology, 46, 165–177.
D.A. Poole, M. Bruck / Developmental Review 32 (2012) 165–180 179
Steward, M. S. & Steward, D. S. (with Farquhar, L., Myers, J. E. B., Reinhart, M., Welker, J., Joye, N., Driskill J., & Morgan, J.) (1996).
Interviewing young children about body touch and handling. Monograph of the Society for Research in Child Development, 61
(4–5, Serial No. 248).
Stuart, R. B., & Lilienfeld, S. O. (2007). The evidence missing from evidence-based practice. American Psychologist, 62, 615–616.
Teoh, Y.-S., Yang, P.-J., Lamb, M. E., & Larsson, A. S. (2010). Do human figure diagrams help alleged victims of sexual abuse
provide elaborate and clear accounts of physical contact with alleged perpetrators? Applied Cognitive Psychology, 24,
Thierry, K. L., Lamb, M. E., Orbach, Y., & Pipe, M.-E. (2005). Developmental differences in the function and use of anatomical dolls
during interviews with alleged sexual abuse victims. Journal of Consulting and Clinical Psychology, 73, 1125–1134.
Trachtenberg, J. (1939). Jewish magic and superstition: A study in folk religion. New York: Behrman’s Jewish Book House.
US Department of Health and Human Services, National Institutes of Health (2007). Understanding clinical trials. <http://>.
US Department of Health and Human Services, Administration for Children and Families, Administration on Children, Youth,
and Families, Children’s Bureau (2011). Child maltreatment 2010. <
Uttal, D. H., Marzolk, D. P., Pierroutsakos, S. L., Smith, C. M., Troseth, G. L., Scudder, K. V., et al (1998). Seeing through symbols:
The developmental of children’s understanding of symbolic relations. In O. N. Saracho & B. Spodek (Eds.), Multiple
perspectives on play in early childhood education (pp. 59–79). Albany, NY: State University of New York Press.
Valenti-Hein, D. (2002). Use of visual tools to report sexual abuse for adults with mental retardation. Mental Retardation, 40,
Walker, A. G. (1999). Handbook on questioning children: A linguistic perspective (2nd ed.). Washington, DC: American Bar
Association Center on Children and the Law.
Walters, S., Holmes, L., Bauer, G., & Vieth, V. (2003). Finding words: Half a nation by 2010. Alexandria, VA: American Prosecutors
Research Institute. <>.
180 D.A. Poole, M. Bruck / Developmental Review 32 (2012) 165–180

Leave a Reply

Your email address will not be published. Required fields are marked *