Memory is a mental process that
helps us remember and recall information and events that happened in the past.
It’s accuracy is however questionable, it is thought to be subjective to
distortions and recconstructions caused by post event information or schema
filtering. This is important to consider especially in cases of eye witness
testimony. Elizabeth Loftus questioned the reliability of eye witness testimony
after an innocent man was arrested because he resembled the real criminal
(Loftus, 1980). In 1974 Loftus and Palmer conducted a study to test the
reliability of memory.
The aim of the study (Loftus and
Palmer, 1974) was to see how changing leading words in questions affected
memory recall. The original study was conducted on forty five american
students, who were seperated into five groups each consisted of nine students.
The study had an independant measures design laboratory experiment in which
each group was shown a set of seven videos of car crashes in random order.
After showing them the film the participants were given a questionarre, in each
of the groups the leading question had a verb of different intensity. The
question was phrased „About how fast were the cars going when they *verb* eact
other?” and the questions were “smashed’,
‘collided’, ‘contacted’, ‘bumped’ and ‘hit’. The findings of this study
showed that smashed had the highest mean speed estimate while contacted had the
lowest, Loftus and Palmer(1974) argued that this could be because of an actual
memory distortion or that the participant tried to adjust their answer to the
strength of the verb in the leading question.
aim of our study was to see how using
different verbs in leading questions on affected the estimated speeds given by
study was an experiment, which gave us the cause and effect relationship of the
variables, with an independent measures
design. The independant measures design was chosen because it is less time
consuming and it eliminates order effect.
The independant variable was the weight
of the verb, meaning the difference in the significance of the verbs, used
in the leading question, and the
dependant variable was the estimated speed given by the participants. The
study had two conditions,
experimental and control. In the first, experimental
condition the verb used in the
question was „smashed” and in the
second condition the verb was
„contacted”. To control extraneous variables we used additional questions in the survey to mask the leading question, to avoid demand charecteristics the participants were not introduced with the real aim of the study in the briefing,
to control situational variables the
participants were asked not to move or speak during the experiment
and we made sure to keep quiet to not disturb the participants.
The participants were given
consent forms to sign before the beginning of the experiment, they were briefed and told they have the right to withdraw from the experiment
at any time, they were also debriefed
of the real aim of the experiment and explained the original study as well as
the theoretical background of the study.
study was conducted on eighteen high
school students, all attending thrid grade. Fifteen of the students aged seventeen and three sixteen, twelve of them were females
and six males. The participant
sample was an oppurtunity sample
because it was the most convinient for us, it was the most time saving since
the participants attend the same school as us and it served as a control over
researcher bias in choosing each participant specifically. The partiipants were
all of the same nationality, and the majority had the same mother tounge
however two of the participants spoke different languages. The participants
were randomly allocated into the two
condition groups, by drawing numbers from a bag, to ensure that each one had
the same chance of being put in each of the conditions.
materials needed was a computer connected to a projector, a video of a car
crash, two sets of questionarres (appendices i), consent forms (appendices ii),
a briefing form (appendices iii) and a debriefing form (appendices iv).
experiment was conducted in a group of four: Din, Zerina, Sara and me. Firstly,
the eighteen participants were gathered in a big classroom. After they have
chosen where to sit, the experimenters introduce themsleves and the
participants are read the briefing, after which they are given out consent
forms to sign.
The participants, then, draw numbers
(1 or 2) from a bag in order to seperate them into condition 1 and 2. One
condition is on the left side of the room while the other is on the right side
of the room. After they are seated they watch the car crash video once,
projected on a panel, when the video is finished they are handed out the
surveys to fill out. Two examiners hand out the surveys, for it not to be
suspicious why one examiner has two stacks of the supposed same surveys.
The surveys are collected, and after
that the participants are debriefed of the nature of the project, they are
given an email adress on which they are able to contact the researchers if they
wish to withdraw from the experiment or if they have any additional questions.
They are told that they would recieve the results during the following month.
raw data from the survey were asembled into two tables (see appendices v). The
reason meadians were chosen as the central tendency is because the data in the
contacted condition is sparsely distributed and mean would not be a fitting
measure, while in the smashed condition the mean and the median show the same
center tendency value, but since there are outliers in the data the median is
less affected by them than the mean. Standard deviation was chosen since it is
more easily intrepreted than the variance, and it is a good measure of
dispersion to show how spread out the data is. From the calculation (see
appendices vi) the following were chosen for describing the data: