magazine issue 1 2015 / Issue 24

Children are poor witnesses. Or are they?

• written by Nathalie Brackmann, Henry Otgaar, Melanie Sauerland & Harald Merckelbach
• edited by Jay Wood

So, based on the above, we find that suggestion-induced false memories are more likely to occur in children than in adults, whereas spontaneous false memories are more prevalent among adults than children. A series of experiments showed that even suggestion-induced false memories can increase with age (Otgaar, Howe, Smeets, Brackmann, & Fissette, 2014). This developmental reversal in suggestion-induced false memories occurs when participants are misled about related but non-presented details that share the same underlying gist representation. In these experiments, younger children, older children, and adults were presented with a video of a robbery. During the video, several details were presented (e.g., the culprit), but several related details were left out (i.e., a weapon). Then, participants received misinformation about these missing related details. When using this procedure, adults and older children were more likely to retrieve the gist information and to accept the related misinformation than younger children. This shows that even forensically-relevant conditions that originally fuelled the assumption of children being exceptionally susceptible to false memories can lead to significant age increases in false memories.  

Consequences

Whereas early findings suggested that children’s accounts are generally more vulnerable to memory distortions, more recent research predicts that not children but adults are particularly susceptible to forming false memories in highly meaning-connected situations. This new view has been supported by work using the DRM paradigm, showing that adults more often than children report the critical, non-presented words that are related to the presented word list. An age increase in spontaneous false memories has also been demonstrated in more ecologically valid contexts. In eyewitness identification research, for example, younger children were found to be less likely to misidentify an innocent bystander as being the thief than older children (Ross et al., 2006). Furthermore, evidence indicating that adults can have higher false memory rates than children even for false memories induced by suggestion is accumulating (Connolly & Price, 2006; Otgaar, Howe, Smeets, et al., 2014).    

What are the consequences of these findings? Should child testimony be refused or favoured as evidence in court when the statement of an adult is available? Neither the idea that children always tell the truth nor the one that child testimony is generally afflicted by memory distortions seem to be adequate rules of thumb. On the one hand, children are capable of giving highly accurate testimony (Bidrose & Goodman, 2000), but on the other hand, they might give faulty accounts due to the formation of false memories. Even though faulty eyewitness accounts are possible in children, these findings do not imply that false memories are absent in adults. As pointed out, under conditions in which adults display greater knowledge of a standard scheme of occurrence than children, false memories are more likely to be present in the adult age group. A thorough analysis of each individual case is therefore imperative to assess the credibility of a child’s statement. It is necessary to establish how and under which circumstances allegations were first brought up and how they developed over time. In light of developmental reversals the widespread assumption of legal practitioners believing that an adult’s statement is nearly always more accurate and trustworthy than that of a child is untenable and needs to be corrected.

Paradoxically, children might be more reliable in situations that are very familiar to adults, but not to children. Especially when there is reference to a unique occurrence and the child makes a spontaneous utterance without previous pressure or influence, their statements should not be left aside. Consequently, in the aforementioned murder case, no external influences provide reason to doubt the 6-year-old girl’s identification of her father as the perpetrator.