Student number: P100568
Instructor: Dr EMMA Meaburn
Subject: Introduction to Research Methods
Schema-driven memory: misremembering
details of everyday subjects due to schemata knowledge
In an attempt to replicate French
& Richards’s watch experiment (1993) which suggests that some everyday
memory failures may be affected by schema-driven memory, an experiment was
conducted on 111 psychology students from Birkbeck University. The students
were presented with a watch with Roman numbers, in which the number four was
represented as IIII, rather than conventional IV. 37 students had to draw the
watch without any previous warning that they would need to draw it from memory,
further 37 students were warned in advance, while the remaining 37 students
were assigned as a control group and had to copy the picture. In the first two groups,
most of the students drew the number four in the IV Roman numerals while the
control group made no mistakes. It was concluded that everyday memory uses the
schema theory to fill in gaps of knowledge that we have previously dismissed as
The schema theory was first proposed
by Jean Piaget (1923). The concept was later popularized by Bartlett (1932) in
a series of experiments in which participants had to recall a Native American fairy
tale (“The War of the Ghosts”) several times over the period of a few
months. Bartlett noticed that the participants remembered the main idea of the story,
but they changed unfamiliar elements to make sense of the story by using terms
more familiar to their own cultural expectations. Bartlett suggested that human
beings possess a set of unconscious mental structures, known today as mental
schemas which produce errors in recall (Bartlett, 1932). Similar research
exploring the mental schema theory found that people were often misremembering
the appearance of everyday objects. Nickerson & Adams (1979), for instance,
asked American participants to draw a picture of a U.S. penny coin from memory.
The participants could only recall three of the eight critical features of the
coin. In a similar experiment conducted by the British psychologist P. E. Morris
(1988), participants were often misremembering the correct appearance of a ten
The schema theory provides the
theoretical basis for French and Richards’s watch experiment (1993). French and
Richards found that participants asked to draw from memory a watch with Roman
numerals on its face drew the number four as IV, rather than the correct IIII.
On the other hand, participants who were simply asked to copy the watch drew
IIII. The current experiment followed the methodology of French and Richards in
an attempt to replicate and extend their research. Subjects drew a watch with
Roman numerals either from memory or while having visual access to the watch. The
findings of this experiment supported the idea that our everyday memory could
be influenced depending on whether or not we are using our schematic knowledge
we have acquired in the past.
It is predicted that participants
drawing the watch from memory will use their schematic knowledge and represent
four as IV, compared to the ones simply copying the picture. It is also
predicted that the participants in the control group will draw four as IIII and
more likely report something unusual about the representation of number four on
the watch face.
experiment had a mixed design to avoid practice effect. The independent
variable was one of the three memory conditions the participants were assigned
to. The dependant variable was whether the participants drew IIII or IV. The
level of measurement is nominal. The participants were split into three groups:
group 1 and 2 experimental and group 3 control group. All three groups were
given the same picture, during the same time of the day.
Computers and a web link to the test
were provided by the university.
A picture of a watch with Roman
111 participants took part in the
experiment. Each participant was randomly allocated to one of the three experimental
conditions, 37 per condition. All participants were psychology students at
Birkbeck University. Age and gender data was not collected. All participants
were naive to the experiment.
Following recruitment, participants
were led to a lab room at the university, where every participant was provided
with a desk where they could sit and perform the test online via a web link
provided by the experimenters.
Participants were then allocated to
one of three groups (1, 2 or 3), unaware of the task ahead. After that, they
were provided with the web link to an online portal asking students for their
consent, age, gender and the number of the group they had been assigned to.
Each group number corresponded to a relevant allocated condition as follows:
1) Surprise memory condition.
37 participants were unaware of the task they had to perform. All they were
told was that they would see a picture of a watch and should examine it closely
for one minute. After one minute the picture disappeared from the screen, and
participants were instructed to draw the watch as accurately as possible on a
piece of paper. They were given five minutes to perform the task, after which
they were asked if anything seemed unusual about the watch face, and they
responded with Yes or No. Subsequently, participants were asked to look at
their drawing to check if they drew the four as IIII or IV (or something else).
In the final step of this condition, the participants had to answer how they
had drawn the four by choosing “1 for IIII, 2 for IV, or 3 for something else”.
2) Forewarned memory condition.
In this case, 37 participants were warned they would have one minute to examine
the watch, after which they would be asked to draw it from memory as accurately
as possible. They were given five minutes to draw the watch, after which they
were asked if anything seemed unusual about the watch face, and they responded
with Yes or No. Subsequently, the participants were asked to look at their
drawing to check if they drew the four as IIII or IV (or something else). In
the final step of this condition, the participants had to answer how they had
drawn the four by choosing “1 for IIII, 2 for IV, or 3 for something else”.
3) Copy condition.
Group 3 was used as a control variable. The remaining 37 participants were
informed they were going to see a picture of a watch, and were given five
minutes to draw the watch while the picture remained in full view for the whole
period. Similarly to group 1 and 2, the participants had to record whether they
noticed anything odd about the watch face, and then to check and record how
they had drawn the number.
All subjects were debriefed after the
completion of the experiment.
The number of times four was
represented as IIII or IV from each assigned condition were calculated for each
participant. The scores for each condition are shown in Table 1. Furthermore,
the number of times participants reported something strange in the watch face
was recorded in Table 2.
data were analysed using the Chi Square test of association. There was a strong
correlation between the “Surprise memory” and “Forewarned memory” condition with
the number four represented as IV (Refer to Table 1). On closer observation,
the calculated Chi Square value (x2=34.98, df=2, p<0.05) massively exceeds the critical value of Chi Square (x2=5.991, df=2). This means that there is a significant association between the assigned condition and the figure drawn, with the majority of those allocated to surprise memory or forewarned memory drawing the figure IV, and those in the copy condition drawing mainly IIII. Table 1. Correlation between allocated condition and figure drawn Figure drawn IIII IV Assigned Condition "Surprise memory" 5 32 "Forewarned memory" 11 26 "Copy" 29 8 Total 45 66 Additionally, the majority of participants from both memory and copy condition claimed to have found nothing odd in the watch (Refer to Table 2). The copy condition × found oddity did not reach significance (x2=5.870, df=1, p<=0.005) as the calculated Chi square value does not surpass the critical value (x2=3.842, df=1, p<=0.05). Table 2. The number of participants who reported oddity and their condition Found oddity No Yes Condition "Memory" 67 7 "Copy" 27 10 Total 94 17 Discussion The results are consistent with the first experimental hypothesis: those who were assigned to one of the two memory conditions drew the figure IV and those assigned to the copy condition mostly drew IIII. This could be explained through the mental schema theory. The participants from the memory conditions were using their mental schematic knowledge of Roman numerals when drawing numbers on the watch from memory. This observation supports mental schema theory and expands upon previous research on recognising features of everyday objects by Nickerson & Adams (1979) and Morris (1988). The second hypothesis was not supported by the research findings as most participants from the copy condition did not find anything irregular about the watch. Perhaps the schema theory could still explain this: participants from the copy condition were not warned to look out for something odd in the watch in advance. However, there is another possible explanation for the above finding that also requires consideration. A possible confounding variable could be time of the day. The experiment was conducted at 7:45 in the evening, this means that the participants' general alertness, as well as the chance to find something odd in the picture, could have been decreased compared to earlier hours of the day. Unfortunately, it is not possible to assess the extent of "participant's alertness" in this experiment. Future work should therefore check on this possibility, and, if possible, warn the control group to look out for something odd in the watch while copying it. The findings of this report generally supported the findings of French and Richards (1993) and the mental schema theory in producing errors in everyday memory. What was found is that memory condition did influence the way participants represented the number four by dismissing their knowledge of Roman numbers. The experiment, however, did not establish a correlation between memory conditions and oddity perception. References Baddeley, A. (1990). Human Memory: Theory and Practice. Hillsdale, NJ: Lawrence Erlbaum Associates. (See pages 335-347.) Bartlett F (1932) Remembering: A study in experimental and social psychology. Cambridge: Cambridge University Press. French, C. C. Richards, A. (1993). Watch this! An everyday example of a schema-driven error in memory. British Journal of Psychology, 84, 249-253. Morris, P. E. (1988). Expertise and everyday memory. In M. M. Gruneberg, P. E. Sykes (Eds.) Practical aspects of memory: Current research and issues, Vol 1. Chichester: Wiley. Richards, A., French, C.C., & Harris, P.R. (1996). Does watch-watching make you watchwise. Memory, 4, 49-58. Richards, A., French, C.C., & Harris, P.R. (1998). Mistakes around the watch: Errors in memory for the orientation of numerals. The Journal of Psychology, 132, 42-46.