Proactive interference: previously learned material interferes with new material.
Retroactive interference: newly learned material interferes with the old one.
Both types of interference act on both short-term and long-term memory.
Brown and Peterson and Peterson and Peterson tested to see whether the interference or decay account
was right. In the first condition in one of their experiments, participants were asked to count backwards
by a certain range from a certain number and report a string of letters at various time intervals while
counting backwards. The participants in the second condition did not have to count backwards and
were also asked to recall a string of letters at various time intervals. People in the interference condition
started losing information after about 18 seconds, whereas those in the no-interference condition started
losing it after about 30 seconds. This shows that both decay and interference contribute to forgetting; if
there was no decay, people in the no-interference condition would never have forgotten the string of
letters; if there was no interference, there would have been no significant difference between the
groups. The difference in time that is observed was attributed to rehearsal of information by people
who were in the no-interference condition; we can suppose that the string was switched from short-
term memory to long-term memory.
Serial position effect: Participants were asked to learn a list of items and then write down as many of
the items from the list in no particular order. It was shown that people recalled best items from the end
of the list, followed by the items at the beginning of the list, finally followed by the items in the middle
of the list. The phenomenon of best remembering things heard last is called the recency effect and the
phenomenon of remembering things heard first pretty well is called the primacy effect. The theoretical
information goes as follows: information heard shortly before recall will still be in short-term memory
and can be remembered easily; information heard at the beginning is rehearsed and remembered pretty
well; information in the middle is not encoded because we are busy rehearsing the first items.
Can we eliminate the primacy effect without affecting the recency effect? If primacy truly reflects
rehearsal and storage of information in long-term memory, we should be able to eliminate it if we
prevent rehearsal. For instance, we can present words individually for very short periods of time; this
has been shown to eliminate the primacy effect. When we prevent rehearsal, items at the beginning of
the list are remembered about as little as items in the middle, whereas items at the end are still
remembered very well.
Can we eliminate the recency effect but spare the primacy effect? If recency really reflects information
available in short-term memory, we should be able to eliminate it by preventing rehearsal at the end. By
presenting an interfering task right after the last items on the list, we can eliminate the recency effect.
E.g.: count forward by 7 starting with 793. How many increments until X? This has been shown to
eliminate the recency effect but not the primacy effect. This shows us that the recency and primacy
effects are due to separate processes, and are like separate entities.
Atkinson and Shiffrin proposed a model that reflected exactly what we've been talking about; rehearsal
transfers information to long-term memory; information that is not rehearsed decays with time.
However, some studies have shown that information does not necessarily have to go through short-term
memory. Patient E.e had impaired short-term memory but could still form new long-term memories;
showing that it is possible for sensory information to go directly from the sensory buffer to long-term
memory. Baddeley's model of short-term memory refers to this type of memory as working memory; it is a much
more dynamic model. This person refers to short-term memory as a limited capacity s