Do child witnesses' confidence levels show how accurate they are?
Despite legal beliefs that child eyewitnesses are unreliable, older children can use confidence ratings to signal the accuracy of their identifications.
Source
Child witness expressions of certainty are informative
Inspect source passages
Do child witnesses' confidence levels show how accurate they are?
These passages occur verbatim in the stored paper text. They support the points listed below; they do not verify every statement in the explanation.
2,205 children in the study
The final sample was therefore 2,205 children (aged: 4–17 years, M = 8.08 years, SD = 2.72 years; sex: 49% female; ethnicity: 81% Caucasian, 9% South Asian, 6% Mixed, 2% Black, 1% Chinese, 1% Other).
6 faces in the lineup
For each confidence level, we calculated suspect ID accuracy using the formula (correct ID rate)/(correct ID rate + ∼false ID rate), where ∼false ID rate refers to the estimated innocent suspect ID rate which is calculated by dividing filler IDs from target-absent lineups by the number of lineup members (6), and then dividing that by the number of target-absent lineups ().
What they did
The researchers reanalyzed historical data from previous studies. They then conducted an original experiment with 2,205 children between the ages of 4 and 17. The children watched a video of a man completing a normal daily routine. Afterwards, they were asked to identify him from an interactive lineup containing 6 faces, and they recorded their confidence before and after making their choice using a scale with 5 levels.
What they found
Memory discrimination improved with age. In target-present lineups, the overall correct identification rates were .32 for the youngest group, .43 for the middle group, and .55 for the oldest group. Using specialized analysis, the researchers found a positive relationship between confidence and accuracy for children in middle childhood and late childhood.
The limits
What it doesn't show
This research does not show that children are perfectly accurate; even older children showed some overconfidence and did not achieve complete accuracy at high confidence. Additionally, the study used non-violent videos, so it does not demonstrate how children perform under the stress of real-world crimes. Finally, the experiment used a brief distraction task, meaning it does not address how delay affects memory confidence over longer periods.
Key terms
- Confidence-Accuracy Characteristic (CAC) analysis
- An analytical technique that plots eyewitness accuracy against confidence levels, focusing only on suspect identifications while excluding filler choices.
- Lineup
- A police identification procedure where a witness is shown a suspect alongside innocent individuals to see if they can identify the culprit.
- Filler
- An innocent person placed in a police lineup who matches the general description of the suspect.
- Target-present lineup
- A lineup that contains the actual culprit of the observed event.
- Target-absent lineup
- A lineup that does not contain the actual culprit and instead contains only innocent people.
- Metacognitive monitoring
- The ability to evaluate and assess the accuracy of one's own cognitive processes or memory.
Flashcards
0 of 35 answers reviewed
Research intelligence for this paper
See its role on concept claims, tensions it is part of, placement history, and related discoveries.
Quiz yourself
In the context of a police lineup, what are "fillers"?
Common questions
Why does the legal system believe child witnesses are unreliable?
Older studies used correlation methods that underestimated the link between confidence and accuracy, leading to the assumption that child certainty is useless.
How did the interactive lineup work?
Children used a computer mouse to rotate 3D facial models on a vertical axis, allowing them to examine the faces from multiple angles.
What was the scale used to measure confidence?
Children used a child-friendly water-cup rating scale, where varying amounts of water represented how sure they felt.
Did the youngest children perform poorly because they didn't understand the task?
No, pilot tests showed that even young children understood the instructions, indicating the lack of confidence-accuracy tracking reflects developmental limits rather than task confusion.
More on Metacognition