Charness (1997)
When can acquired knowledge compensate? Charness distinguishes between two types of knowledge acquisition: A and D. The A-system "acts as a frequency-driven event encoding system that seems to work with minimal attentional resource." The D-system, by contrast, represents effortful learning. Age differences are minimal for A-system learning, but substantial for D-system learning.
Knowledge-Computation Tradeoff
How can people use knowledge to circumvent declines in basic processing efficiency? Charness delineates three features of knowledge: "it consists of stored information that can be activated in a timely fashion in order to generate an appropriate response." Hence, there are two major dimensions of knowledge: accuracy and speed. The latter is the dimension that may prevent complete compensation by older adults.
Charness explains that knowledge compensates for deficits in processing efficiency "by replacing computation (sequences of effortful retrieval) with relatively effortless retrieval." For example, given the problem 20 x 20, most people can quickly retrieve the answer of 400. Give the problem to a fifth grader, however, and he or she might have to compute the problem. Processing efficiency can, however, surpass knowledge, as is the case with Deep Blue. Is more knowledge always better?
Access, Acquisition, and Persistence
It seems reasonable to predict that age-related slowing in access might be attributable to age-related increases in knowledge. That is, older adults may have more knowledge to search. However, Charness estimates that only 6-12% of slowed access can be explained by increases in knowledge. Another cost of more knowledge is proactive interference, which occurs when old information interferes with new learning. Nevertheless, knowledge can also have a facilitative effect if organized appropriately. That is, it’s easier to learn new information if it can be related to and integrated with what’s already known. Finally, because procedural knowledge is automatically activated under the right conditions, an old solution might be applied when a new one is more efficient (this phenomenon is called functional fixity). Priming may also activate irrelevant knowledge, and loss of inhibition may prevent activation of irrelevant knowledge structures.
Compensation and Maintenance
Experts can readily access domain-relevant knowledge. Ericsson and Krampe suggest that preservation of high levels of performance is attributable to sustained deliberate practice. What does deliberate practice preserve? Charness states, "One of the distinguishing features of experts is that they maintain high amounts of current deliberate practice, thereby maximizing the ready accessibility of the information." Frequency—how often a stimulus is encountered--and recency—how recently a stimulus has occurred--are two factors that may affect long-term retention and retrieval. Frequency of information seems to be the key factor in retention, but recency influences retrieval. Charness argues that deliberate practice supports both. Estimates of deliberate practice do, however, seem to decline with age. Thus, "Older adults are less likely to have all their knowledge in a readily accessible state."
Comments and Questions
How does practice affect accessibility of knowledge? To put it another way, what are the factors that influence retention of and access to knowledge in long-term memory. Charness pointed to two such factors: recency and frequency. Frequently encountered information is better retained, and recently retrieved information is more easily retrieved in the future. Thus, current practice may influence LTM processes. The literature on long-term memory will provide more cues about how practice might preserve performance. For example, what happens to production rules if they aren’t executed frequently?