The effect of misleading information and leading questions on eyewitness testimony

How susceptible is eyewitness memory to being mislead by questions or false information?

The effect of leading questions
Loftus & Palmer (1974)
The aim of this study was to investigate the effect of leading verbs on eye witness accounts of a car crash. All participants watched a film in which two cars crashed in to each other. They were then randomly allocated to one of five groups and asked to answer some questions about the film. Each group had one question that differed slightly over the five conditions of the experiment,

“About how fast were the cars going when they collided with each other?”

(The important verb in the question was changed to either contacted, collided, smashed, hit or bumped.)

In the ‘smashed’ condition participants gave a significantly higher speed estimate (40.8 mph) compared with the contacted, collided, hit and bumped conditions (31.8, 34.2, 38.1 and 39.3 mph respectively). This showed that the type of question an eyewitness is asked can have a distorting effect on their memory of an incident. In this case the verb used in the question led participants into making a different type of speed estimate.

Loftus & Palmer followed also wanted to investigate just how widely the type of questioning asked could affect participants’ memories of an incident. They followed their initial study with a second one in which participants were asked the same leading questions again (this time only hit and smashed were used, and there was also a control condition with no speed related question) but were also asked whether or not they had seen any broken glass.

“Was there any broken glass at the scene?”

47% of participants in the ‘smashed’ condition reported seeing broken glass compared with 16% in the ‘hit’ condition and 13.5% in the control condition. There had in fact been no broken glass in the film scene, but participants’ memories were again led into incorrect recall by the simple verb they were asked. This shows that an eyewitness can be led into providing incorrect recall of events by the type of questions asked related to the event.

As this was a laboratory experiment Loftus & Palmer were able to establish an extremely strong causal link between their independent variable (the verb used) and their dependent variable (the answers participants gave to the questions), and could be relatively certain that most if not all extraneous variables were controlled for. However laboratory experiments lack ecological validity, and this particular study does not fairly represent events in real life as a real life incident would probably be witnessed directly and without warning, and not on a film clip in a comfortable laboratory.

The participants used in the study were all American university students which means that the results lack population validity outside that of American university students, in other words they cannot be generalised to the majority of people in the world.

The effect of misleading information
Misleading information, also known as after the fact information, can change the memory of an eyewitness by providing information that becomes incorporated in to the memory of the event even though it was not present at the time of the event.

Loftus (1975)
Loftus demonstrated the power of misleading information in a laboratory experiment in which participants were shown a film clip of a car accident. After watching the film participants were asked either, “How fast was the white sports car going when it passed the barn while traveling along the country road?” (the barn condition) or “How fast was the white sports car going while traveling along the country road?” (the control condition).

“Did you see a barn?”

Even though there was no barn in the film clip, over 17% of participants in the barn condition claimed one week later that they had seen a barn compared with only 3% in the control condition. The simple question about how far the car was travelling when it passed the barn was enough to add a barn to memories of 17% of the participants in the study. It is not hard therefore to see how a question that adds new information may change the memory of a witness to a crime!

As with all laboratory experiments, there was good control of extraneous variables which allows a causal link to be established between the question type and the answers participants gave to the questions. The study was however low in ecological validity as the task was unrepresentative of everyday events (i.e. a film rather than a real life incident), and it was low in population validity as the participants were all American college students.

What about blatantly incorrect information?
When misleading information is provided that is blatantly incorrect, witnesses are generally more resistant to being misled and tend to stick to the events they remember witnessing. Loftus (1975) showed participants a set of slides of a handbag robbery. Immediately after seeing the slides participants were asked what the colour of the handbag was, with the result that 98% correctly answered that it was a red handbag. Participants then read an account given by a professor of psychology who witnessed the incident - one version of the witness account had many minor incorrect facts and the another version stated that a brown handbag had been stolen. Almost all of the participants who read the brown handbag account resisted the incorrect information and stuck to their memories of a red handbag.

A Level exam tips
Answering a 12 mark question (PSYA1 AQA A specification)
Outline and evaluate research into the effects of misleading information on eyewitness testimony.
6 AO1 marks - describe the research. This page details a study by Loftus & Palmer (1974) into leading questions, a study by Loftus (1975) into misleading after the fact information, and a study by Loftus 91975) into blatantly incorrect information.

6AO2 marks come from evaluating the research. Discuss the studies in terms of them being laboratory experiments that have the strengths of control of extraneous variables and good causal inference, but the weaknesses of low ecological validity and low population validity. Explain why these are strengths or weaknesses for the studies.