The past continuously besets our ability to act flexibly in the future; habits grow strong, automaticity takes over and the mind wanders. Before you know it, you’ve forgotten to stop for milk on your regular commute, neglected to go to your dentist appointment, or merely “lost track” of what you were doing. These kinds of things are often studied in cognitive control laboratories, usually in the context of a task where there is particularly strong “proactive interference” from previous habits (for example, the habit to read color-words instead of naming ink-colors in the Stroop task).
A recent article by Robert Kail uses a more subtle form of proactive interference to study the growth of cognitive control in childhood. Initiallly, age-related reductions in proactive interference were thought to reflect changes in rehearsal strategies (i.e., whether you are visualizing the memoranda, or repeating them to yourself subvocally), but more recent theories emphasize the importance of “inhibition” in reducing interference. For example, young children are less able to intentionally forget things in a directed forgetting paradigm (discussed yesterday), which suggests to some that they may be less able to cognitively suppress items in memory.
On the other hand, Kail reviews evidence from several studies showing that vulnerability to interference does not change throughout childhood. To examine this contradictory finding in more detail, Kail meta-analyzed 26 studies of children performing a test of free recall. Kail showed that when the results from these studies are pooled together, they do in fact show that proactive interference declines with age (and that it increases with the total number of items studied).
To overcome some of the limitations with meta-analyses (such as the “closed-drawer problem” in which null findings are not usually published), Kail then conducted his own study with 100 children between the ages of 9 and 12, and 25 undergraduates. Each subject was presented with three familiar one-syllable words, asked to count backwards from a three-digit number, and then asked to recall each of the one-syllable words. This process was repeated four times in a row, after which subjects engaged in a different 5-minute processing speed task (essentially the total time required to complete several choice RT tasks). This entire process was than repeated for a total of 4 times.
The results showed that the amount of proactive interference (i.e., the degree to which recall degraded over successive trials) was largest for the youngest children, and decreased with age. Structural equation modeling showed that the data were consistent with a theory where age-related differences in interference are completely explained by differences in processing speed (which are themselves affected by age), without the need for a specific relationship between age and interference.
Kail concludes that although previous individual studies had not found age-related changes in proactive interference, this trend is actually apparent if enough data is analyzed, and is also apparent even with older children. Kail interprets “escape from interference” as the result of improvements in executive control and processing speed. Alternatively, age-related differences in working memory span could reflect the differences in interference, as argued extensively in Unsworth & Engle’s recent article.