The roles of schemata in information organization and retrieval
Schemata have been defined as ‘knowledge structures … based on past experience, which
exist at various level of abstraction…’ (Brewer & Treyens 1981, p. 208). These mental
representations are thought to allow for the formation of expectations, facilitate the
understanding of perceptual input, expedite retrieval and guide behaviour (Ghosh & Gilboa
2014, p.105). Bartlett was the first psychologist, who investigated the operations of schemata
in relation to memory processes. In his theory, he postulated that people’s perception and
subsequent recall of new information is heavily influenced by their previous unique
experiences (Groome et al. 1999, p.7). Since people’s knowledge and experiences can vary
greatly, the way in which a particular perceptual input is interpreted can also differ from one
individual to the next. The current essay aims to examine the roles schemas play in
information organization and retrieval by discussing relevant research evidence from the
schema literature.
Bartlett (1932) suggested that memory is heavily influenced by past experiences and pre-
existing schematic knowledge. According to him, remembering is a reconstructive rather than
a reduplicative process, which operates in a cultural context. He maintained that if there was
a conflict between ‘the observed’ and ‘the expected’, schemas would bias reconstructive
processes leading to the production of systematic errors in recall (Eysenck & Keane 2010,
p.401). To investigate the effect of reconstructive processes on recall, Bartlett used different
stimuli- most notably North American Indian folk tales such as ‘The War of the Ghost’-, and
different methods- the two best known ones being the serial and repeated reproduction
techniques. In serial reproduction, an individual is exposed to a stimulus (a story), which
he/she then recalls. The second person will read the first person’s account of the story, and
will then try to recall that version. In repeated reproduction, the same person recalls the story
on different occasions. Bartlett found that his participants tended to recall a changed version
of the tale. However, their distortions were not random, but were directed towards the
creation of a story, which was more conventional, sensible, more in accord with their own
perceptions of reality (Carlson, Buskist & Martin 1997, p.266). Bartlett referred to this process
of relating new information to old schema-based material as ‘effort after meaning’ (Brewer
& Nakamura 1984, p.9). Essentially, the participants tended to rationalize the supernatural
1
, Indian folk tales to make them more readily fit in with their own cultural expectations (Bartlett
1932, p.209).
At the time of the publication of Bartlett’s Remembering, mainstream psychology was
dominated by the behaviourists, who tended to focus solely on stimuli-response relations and
disregard all inner mental processes, which could not be observed directly. It hardly comes as
a surprise, that Bartlett’s theory of reconstructive remembering, to which the concept of
‘schema’ was so central, was not particularly well-received at first. Over the years, many have
regarded the definition of schema to be far too vague, arguing that it has not been defined
precisely enough to enable it to be empirically tested in a reliable fashion. Ghosh and Gilboa
(2014) pointed out that there still seems to be considerable debate among professionals over
what actually constitutes a schema.
Furthermore, Bartlett’s informal way of conducting research- not giving specific
instructions to, and testing his participants in a non-systematic manner- has also been
subjected to much criticism. Bartlett himself admitted that he had deliberately provided his
participants with somewhat lenient instructions, in order to exert minimal influence on their
procedure and keep his experiments as natural as possible. However, Gauld and Stephenson
(1967) argued that the recall distortions documented by Bartlett might have occurred
precisely as a result of his vague instructions, which could have potentially led his participants
to construct rather than to accurately remember the story. To investigate this phenomenon,
the researchers conducted three experiments varying the type of instructions given to their
participants. They found that strict instructions, which emphasized the importance of
accurate recall, greatly reduced the number of errors made. However, these findings do not
necessarily undermine Bartlett’s theory of reconstructive remembering. As Bergman and
Roediger (1999), who have successfully replicated Bartlett’s findings using ‘The War of the
Ghost’, pointed out, Gauld and Stephenson tested their subjects immediately after the study,
while Bartlett tested his participants with retention intervals. Distortions attributable to
rationalization tend to increase over time, which could explain why Gauld and Stephenson
had different results from that of Bartlett’s.
Bartlett’s choice of material has also been criticized by some for not being ecologically
valid. As Roediger, Bergman and Meade (2000, p.117) argued, it is ‘as similar to normal prose
as Ebbinghaus’s … non-sense syllables are to words.’ However, Bartlett chose these stories
2
