Loftus and Palmer evaluation Flashcards
Low in generalisability through the unrepresentative sample.
45 university students from Washington University in experiment1. Results can’t be generalised to other age groups. Students may be less likely to be experience in driving compared to adults who may estimate differently meaning possibly inconsistent results.
There’s a smaller sample in experiment 1 and a bigger sample in experiment 2.
In experiment 1, the sample consisted of 45 participants with 9 in each condition. However, in experiment 2 the sample had 150 participants with 3 groups of 50. Decreases the internal validity as it increases the chance of anomalies skewing data analysis.
Has a standardised procedure in experiment 1.
They used the same videos, same five conditions and the same questionnaire. For example, the videos were 5-30 seconds long, they used 7 questions and has two parts to the questionnaire that included a description. Procedure is standardised which makes it easier to replicate and test for reliability.
Yuille and Cutshall criticises Loftus and Palmer’s research.
Misleading questions had very little effect on participants recall. 10 of the eyewitnesses said there was no broken head light and no yellow quarter panel on the thief’s car. States that in realistic environments and the effect of misleading questions on EWT surprisingly does not affect accuracy. Lowers reliability of the experiment. It also reduces usefulness of applications such as the cognitive interview.
Has positive application to the police to gain accurate details from EWT.
Loftus found that the verb used in the critical question impacted on the response given eg smashed where the students estimated a higher speed of 40.5mph compared to hit which was 34mph. Therefore, police should be aware when interviewing EW’s and consider how questions may impact answers such as using general open ended questions.
Can be used to advise law teams when taking eye witness testimonies.
Found that using leading questions can influence the answer a person will give such as smashed which gave a speed of 40.5 and hit which gave an estimate of 34mph. So, leading questions should be used by law teams. It can be used to discredit witnesses and persuade the jury by fooling witnesses to say what the lawyer wants.
Has low mundane realism.
Participants watched 7 clips of a car accident through a TV screen in an artificial environment. They also completed the questionnaire straight after witnessing. The participants will be less emotionally engaged knowing that the videos weren’t real and their emotions are not as representative as the consequences aren’t that big compared to real EW’s as no one is being convicted.
Consent was taken before the study.
Used a volunteer sample of 45 students who all gave their informed consent to take part. The experiment has high ethical consideration as there was informed consent. However, the results are less likely to be valid due to demand characteristics which makes it less applicable.
Controlled extraneous variables such as order effects.
Used 7 clips of car crashes of which 4 were staged, Each shown for 5 to 30 seconds in different orders which uses counterbalancing to prevent order effects such as practice. Therefore, this ensures participants don’t practice from film to film, this increases the external validity of results.
High mundane realism due to a realistic task recall.
Participants were asked general questions and 10 specific questions in a 2 part questionnaire. Similar to a police interview where they ask you to recall general information then ask specific questions about it, therefore the study shows findings of a realistic task.
