After a week, the participants were all asked a further ten questions, including “Did you see a barn?” The findings of the experiment showed that only 2.7% of group one gave the incorrect answer to the question, stating there was a barn. Whereas 17.3% of group two claimed they saw a barn. This suggests that the original memory of the participants in group two had been altered by later information. Proving that eyewitness testimonies are subject to inaccuracies and can be altered by misleading post-event information, therefore they cannot be heavily relied upon.
Research has also shown that leading questions can affect the accuracy of eyewitness accounts. For instance, in a study where film footage of a car accident was shown participants were asked one of two questions, “Did you see a broken headlight?” or “did you see the broken headlight?” Participants asked about ‘the’ rather than ‘a’ broken headlight were twice as likely to say yes, when in fact there was no broken headlight. Therefore this shows that leading questions can alter an eyewitness’ recall and in turn affect its accuracy.
Loftus has also suggested that eyewitness testimonies are affected by the anxiety of the witness, stating that in an event in which a weapon is present the fear or anxiety caused by the sight of the weapon narrows their focus of attention. For instance, their recall of the physical appearance of the attacker will be less accurate and therefore should not be relied upon in court.
However, there is also research to support the accuracy of eyewitnesses. For instance, blatantly incorrect information tends not to mislead people or distort their memory. Loftus (1979) proved this when she showed participants a film illustrating the theft of a red purse from a bag. The participants then partook in an immediate recall test and 98% remembered the colour of the bag. The participants where then asked to read an account of the event written by a professor of psychology (the title was given in order to suggest that the account was accurate). The account contained the blatantly incorrect information that the purse was brown. In a second recall test, only two participants accepted the blatantly wrong information as truth, with the rest of the participants correctly recalling that the purse was red. This suggests that memory of information that stands out and is particularly noticeable is less subject to alteration, with people being able to ignore new information, due to the fact they know it is incorrect, and in turn maintain their original memory.
Christianson and Hubinette (1993) also found that recall of an event by an eyewitness was far more accurate if they were involved (e.g. a victim), suggesting that people are good at remembering highly stressful situations. In other words, the tests conducted by Loftus may not reflect the true accuracy of eyewitness testimony as the people were not involved in the event and simply participants within the artificial surroundings of a laboratory. They concluded that eyewitness would be far more accurate in real life as those involved would have experienced a highly stressful situation. This is also supported by Yullie and Cutshall (1986), who interviewed 13 real life witnesses to an armed robbery in Canada. The witnesses were interviewed at the time of the crime and again four months later. Despite the fact some misleading questions were used within the recall tests, their recall was very good on both occasions. This suggests that in real life situations eyewitness testimony can be very reliable, more so than in laboratory based studies.
In conclusion, although eyewitness testimonies are subject to inaccuracies I feel that they are not as unreliable as Loftus’ studies suggest. This is due to the fact that the participants within her studies were not emotionally affected by the incidents, or put in a highly stressful situation, thus their recall would not have been as accurate as in a real life situation.