Brief History of Research on Child Witnesses
In the early part of the 20th century, research on children's memory focused
on children as witnesses in court. Most of this research was in Europe,
especially Germany and France; there was very little in the United States
until the 1920s and even then there were only a few studies on the child
witness until the 1980s (Ceci & Bruck 1993, 1995). Although children
in Canada, America, and Great Britain were rarely permitted to testify,
their testimony was allowed in some other European countries, hence the
interest in performing research that dealt directly with children's court
testimony. The general conclusion from this research was that young children
were suggestible and vulnerable to making serious errors in their court
testimony (Ceci & Bruck, 1993, 1995; Goodman, 1994; Wakefield &
Following this early period, little research was done for years and laboratory
studies of children as witnesses were rare until the 1970s. But then the
increases in reports and allegations of sexual and physical abuse led to
changes in the legal system regarding the admissibility of child witnesses'
testimony (Ceci & Bruck, 1993, 1995; Goodman, 1984). In most countries
the requirement for corroboration of children's statements alleging sexual
abuse was dropped. Therefore, children's credibility, their reliability
as witnesses, and their susceptibility to leading questions quickly became
salient issues. Over the past few years interest in this issue has proliferated
and there are now hundreds of articles in the literature addressing these
During the beginning of this recent period, the testimony of young children
was often accepted in the justice system as truthful and false allegations
were believed to be extremely rare (i.e., Faller, 1984; Summit & Kryso,
1978). A number of unsupported dogmas, such as children cannot lie about
sexual abuse and children cannot be "coached" to make erroneous
statements about abuse, gained acceptance in the legal and mental health
communities. Although some researchers acknowledged that suggestive interviews
might cause children to make unimportant errors about peripheral details,
it was claimed that children could not be led to make statements about important,
central events. Several highly publicized cases during this period involving
satanic, ritual abuse allegations resulted in convictions (Nathan &
To our knowledge, the first time expert opinion on the impact of leading
interrogations was given in a court of law was when Dr. Underwager testified
in a sex abuse trial in Winner, South Dakota, in the winter of 1983. The
next year we were the experts in the first of the highly publicized Scott
County cases to go to trial where Dr. Underwager provided similar testimony.
In these cases 25 adults were accused of abusing 40 children in two interlocking
sex rings. The alleged abuse involved animals, ritual abuse, and murder.
This trial resulted in an acquittal and charges were dropped against all
but the first defendant, who had confessed. A later state attorney general's
report (Humphrey, 1985) concluded that the repeated suggestive interrogations
of the children made it impossible to sort out what may have happened.
Following this, a few forensic psychologists and psychiatrists began to
report on their analyses of actual real world interrogations (Coleman, 1986;
McIver, 1986; Underwager, Wakefield, Legrand, Bartz, & Erickson, 1986).
These early reports of actual real world interrogations suggested coercive
questioning could produce serious errors in a child's statements. Then,
as academic researchers became involved in actual cases and reviewed videotapes
of actual interviews, they began conducting studies that more closely approximate
what happens in the real world. These studies demonstrated that children
who are interviewed suggestively can produce false narratives about fictitious
events, including central events (see Ceci & Bruck, 1995 for a review
of this research). These erroneous narratives were often coherent and detailed
and could not be detected as false by professionals. As a result, the current
consensus of scientific opinion has revived the initial caution and concern
about children's reliability and suggestibility. The justice system now
recognizes these data and several of the highly publicized convictions in
the United States and Canada from the 1980s have been overturned.
It is now acknowledged that persistent suggestive questioning can lead children
to provide accounts of events that never occurred, even when they first
denied them. Sometimes the questioning results in the child developing a
subjectively real memory for an event that never happened. (Bruck &
Ceci, 1995; Ceci, 1994; Ceci & Bruck, 1993, 1995; Ceci, Loftus, Leichtman,
& Bruck, 1994). Several conclusions are now generally accepted in the
scientific community (Ceci & Bruck, 1993):
It is extremely important to examine the conditions prevalent at the time
of the child's original report. The way the child was interviewed, including
the use of image-based techniques, must be carefully examined. The conclusions
from the current research are clear-when children are interviewed skillfully
and appropriately and supported and encouraged to tell their story in their
own words, they can provide accurate and forensically useful information.
But when interviewers use suggestive, leading, specific, and coercive questioning
to get the child to confirm preexisting biases about abuse, they risk eliciting
false statements. Therefore, if image-based techniques are used, it must
be in a nonsuggestive way that encourages children to provide details and
a narrative account from their free recall of events.
· First and foremost, contrary to the claims made by some
. . . there do appear to be significant age differences in suggestibility,
with preschool-aged children being disproportionately more vulnerable to
suggestion than either school-aged children or adults (p. 431).
· Our review of the literature indicates that children can indeed be
led to make false or inaccurate reports about very crucial, personally experienced,
central events (p. 432).
· The second major conclusion is that contrary to the claims of some,
children sometimes lie when the motivational structure is tilted toward
lying (p. 433).
· Third, notwithstanding the aforementioned two points, it is clear
that children-even preschoolers-are capable of recalling much that is forensically
relevant (p. 433).