Memory

 Showed participants films of a car accidents and then asked them how fast the car was going. They changed the verb used in the question to see if that had any effects on the estimates of speed.contacted resulted inestimates of around 30mph and smashed in estimates of 40mph.In extension to the experiment, they took two groups (the hit and smashed group) and asked them "Did you see any broken glass?" when there wasn't any. The smashed group were more likely to report having seen broken glass.

• This was a labortory experiment - possibility that participants did not take it seriously, nor were as emotionally aroused as those who had witness a real-life car accident.Evidence from real life studies does not support the view that misleading infomration affects accuracy.

• DEMAN CHARACTERISTICS - participants alter what they say to please the researcher. Loftus tested this idea by offering participants money if they correctly recall details of a car accident. She divided groups into 4 groups raging from no monetary reward to $25 and stressed the importance of correct answeres - despite this, she found that participants still made errors. This illustrates that misleading questions do indeed influence and dsitort memory.

Elizabeth Loftus reported on research carried out by Johnson and Scott (1976) which investigated the weapon focus effect!

Participants heard a heated discussion in another room - in one condition; a man emerged from the room carrying a pen with his hands covered in grease. In the second condition; a man emerged carrying a knife with blood on his hands. When asked to identify the man from 50 photos, those who had seen the man with the pen gave a far more accurate recall than those who had seen the man with the knife.This suggests that anxiety created by the weapon somehow distracts witnesses' attention from other aspects of the incident.

• The study has good validity as participants were not aware that the study was staged.
• ETHICAL ISSUES - that some participants may have been exposed to stress at the sight of the knife.

Loftus and Burns (1982) showed some participants a violent version of a crim in which a boy was shot in the face and found that participants had impaired recall for events leading up to the violent incident.

• Research such as Lotus' and Johnsons and Sctott's could be seen to lack validity as much of it is carried out in an articiial environment, i.e. laboratory. This means that participants do not take the experiement seriously knowing that their EWT does not really matter and that a crime has not really been comitted. This could be a reason why so many laboratory studies show realtively poor recall and why so many witnesses of real crimes show relatively good recall.

A study of real life crime by Christianson and Hubinette in 1993 found that anxiety actually enhanced the accuracy of memory. In this study they questioned people who had witnessed a real bank robbery. They found that the people who had been threatened in some way remembered more than those people that had just be onlookers. Of course, because this was a natural experiment there may have been other extraneous variables that also affected the accuracy of recall, such as age and individual difference, so it may not have been anxiety.