In the study, 18 children (aged 7-8), 20 adolescents (13-14), and 20 young adults (20-29) were shown pictures and asked to decide whether it was a new picture or one they had seen earlier. Some of the pictures were of known objects and others were fanciful figures (this was in order to measure the effects of novelty in general). After a 10-minute break, they resumed the task — with the twist that any pictures that had appeared in the first session should be judged “new” if that was the first appearance in the second session. EEG measurements (event-related potentials — ERPs) were taken during the sessions.
ERPs at the onset of a test stimulus (each picture) are different for new and old (repeated) stimuli. Previous studies have established various old/new effects that reflect item and source memory in adults. In the case of item memory, recognition is thought to be based on two processes — familiarity and recollection — which are reflected in ERPs of different timings and location (familiarity: mid-frontal at 300-500 msec; recollection: parietal at 400-70 msec). Familiarity is seen as a fast assessment of similarity, while recollection varies according to the amount of retrieved information.
Source memory appears to require control processes that involve the prefrontal cortex. Given that this region is the slowest to mature, it would not be surprising if source memory is a problematic memory task for the young. And indeed, previous research has found that children do have particular difficulty in sourcing memories when the sources are highly similar.
In the present study, children performed more poorly than adolescents and adults on both item memory and source memory. Adolescents performed more poorly than adults on item memory but not on source memory. Children performed more poorly on source memory than item memory, but adolescents and adults showed no difference between the two tasks.
All groups responded faster to new items than old, and ERP responses to general novelty were similar across the groups — although children showed a left-frontal focus that may reflect the transition from analytic to a more holistic processing approach.
ERPs to old items, however, showed a difference: for adults, they were especially pronounced at frontal sites, and occurred at around 350-450 msec; for children and adolescents they were most pronounced at posterior sites, occurring at 600-800 msec for children and 400-600 msec for adolescents. Only adults showed the early midfrontal response that is assumed to reflect familiarity processing. On the other hand, the late old/new effect occurring at parietal sites and thought to reflect recollection, was similar across all age groups. The early old/new effect seen in children and adolescents at central and parietal regions is thought to reflect early recollection.
In other words, only adults showed the brain responses typical of familiarity as well as recollection. Now, some research has found evidence of familiarity processing in children, so this shouldn’t be taken as proof against familiarity processing in the young. What seems most likely is that children are less likely to use such processing. Clearly the next step is to find out the factors that affect this.
Another interesting point is the early recollective response shown by children and adolescents. It’s speculated that these groups may have used more retrieval cues — conceptual as well as perceptual — that facilitated recollection. I’m reminded of a couple of studies I reported on some years ago, that found that young children were better than adults on a recognition task in some circumstances — because children were using a similarity-based process and adults a categorization-based one. In these cases, it had more to do with knowledge than development. I’ve appended the two reports below.
It’s also worth noting that, in adults, the recollective response was accentuated in the right-frontal area. This suggests that recollection was overlapping with post-retrieval monitoring. It’s speculated that adults’ greater use of familiarity produces a greater need for monitoring, because of the greater uncertainty.
What all this suggests is that preadolescent children are less able to strategically recollect source information, and that strategic recollection undergoes an important step in early adolescence that is probably related to improvements in cognitive control. But this process is still being refined in adolescents, in particular as regards monitoring and coping with uncertainty.
Interestingly, source memory is also one of the areas affected early in old age.
Failure to remember the source of a memory has many practical implications, in particular in the way it renders people more vulnerable to false memories.
Children outperform adults in memory study
An example of the perils of knowing too much! — under specific conditions, young children can beat most adults on a recognition memory test. The study compared young children (average age 5 years) with college students. Without being told what was being tested, participants were shown pictures of cats, bears and birds. Some of them were first shown a picture of a cat, and told that it had “beta cells inside its body”. They were then shown other pictures, and asked whether these animals also had beta cells. After this, they were shown other pictures, and asked whether they had been shown them before. The children were accurate on average 31% of the time; the college students only 7% of the time. The researchers suggested the reason was because the children used similarity-based induction: when asked whether each pictured animal had "beta cells", they looked carefully to see if the animal looked similar to the original cat. On the other hand, the adults used category-based induction: once they determined whether the animal pictured was a cat or not, they paid no more attention. Thus, when they were tested later, the adults didn't know the pictures as well as the children. A subsequent study taught the children to use category-based induction. Their performance then dropped to the level of the adults. Another study in which participants were simply shown the pictures of the 30 animals and told to remember them for a recognition test, found adults were accurate 42% of the time, compared to only 27% for the children.
Too much knowledge can be bad for some types of memory
Following on from an earlier study reported last year, in which children were found to have better memories than adults in certain circumstances, researchers have found that adults did better remembering pictures of imaginary animals than they did remembering pictures of real cats. The reason has to do with the effects of categorization. While categorization is often vital, it can lead people to ignore individual details. The trick is to know when it’s important to categorize and when it’s better to note specific details. The new study added to the earlier findings by showing that there is a gradual decrease in recognition memory from children to adults, rather than an abrupt change in the way people see the world. Moreover, the difference in how adults and children perceive and remember objects is not a developmental difference, but one caused by differences in knowledge. Adults performed like children when shown imaginary animals.