Factors affecting the accuracy of Eye-Witness testimonies - Memory Psychology

An eyewitness testimony is…evidence provided in court by people recalling details of an event they have witnessed. This includes identification of perpetrators and details of the crime scene ect.

Factors affecting the accuracy of eyewitness testimonies: misleading information

Misleading information is..any information that ‘leads’ you into giving a particular response.

A leading question is..a question that prompts or encourages the answer wanted.

Why do leading questions affect eyewitness testimonies?

The response bias explanation suggests that the wording of questions doesn’t have an effect on participants’ memories, but influences their answer. For example, Loftus and Palmer, experiment 1, (1974): when participants were asked a leading question with the word ‘smashed’, opposed to ‘hit’, ‘bumped’, ‘collided’ or ‘contacted’, they estimated a high travelling speed of the car – suggesting that this more ‘dramatic’ verb encouraged them to do this.

The substitution explanation suggests that the wording of a leading question changes the participants’ memory.
For example, Loftus and Palmer, experiment 2, (1974) found that participants who had heard the word ‘smashed’ where more likely to report seeing broken glass than those who didn’t. (there was no glass!)

Research: Loftus and Palmer 1 (1974)

Students were shown films of traffic accidents, and given a questionnaire afterwards. There was one leading question included: “about how fast were the cars going when they hit each other?”. The verb ‘hit’ was changed for each of the 5 conditions (hit, contacted, bumped, collided, smashed).
The mean estimated speed for each of the conditions was calculated and it was found that those who had the verb ‘smashed’ had the highest estimated average speed of 40.8mph, whilst those with the verb ‘contacted’ estimated a speed significantly lower: 31.8mph.
This suggests that the leading question biased the eyewitness’ memory of the event – providing support for the response bias explanation.

Evaluation:

P: laboratory experiment
E: can be easily replicated, as any film clips can be easily used again with other participants, allowing another researcher to check the consistency of results.
E: experiment is reliable and accurate.

P: independent measures design
E: no order effects and therefore the results won’t be affected by factors such as becoming bored or fatigued
E: this increases the validity of the results, meaning they can be generalised

 

P: they watched ‘films’
E: this is different from actually first-hand witnessing the event and so the participants may lack the emotions, (like stress) they may have had if witnessing it in real life – and emotions may have influence on memory.
E: leads to a lacking of validity as artifical tasks like this don’t tell us much about real-life scenarios.

Research: Loftus And Palmer 2 (1974)

In a second study by Loftus and Palmer, they had 150 student participants, who were shown a short film of a multi-vehicle car accident, and then were asked questions about it. The participants were spilt into 3 conditions: condition 1 were asked ‘how fast were the cars going when they hit each other?”; condition 2 were asked the same question but with ‘smashed’ opposed to ‘hit’, and the final condition were asked nothing involving speed.
One week later, all groups returned and were asked if they had seen any broken glass (even though there was none!)
It was found that participants who had heard the word ‘smashed’ were more likely to report seeing glass – providing support for the substitution explanation.

Evaluation:
P: participants were university psychology students who knew they were in an experiment
E: demand characteristics may have affected the results as they may have been looking for clues as how to behave/respond, and social desirability bias may have lead them to giving results they thought were wanted by the researcher
E: lacks validity and therefore shouldn’t be generalised to real life situations

P: participants were university psychology students
E: unrepresentative of the population; not only students will witness car accidents, and they may be less experienced drivers and therefore may be less confident in their ability to estimate speed
E: low population validity and thus, lacks generalisability.

---------

Post event discussion

Post event discussion is...when witnesses of a crime discuss details of the event, following the incident.

Why does post-event discussion affect eye witness testimonies?

When co-witnesses to a crime discuss it with each other, their eye witness testimony may become contaminated. This is because they may combine (mis)information from other witnesses with their own memories.

Supporting Research: Gabbert Et Al. (2003)

Gabbert et al (2003) studied participants in pairs – all participants watched a video of the same crime, but filmed from different points of view. Therefore, one member of the pairs could see elements that the other could not eg. only one participant could see the title of the book being carried by a young woman. The participants discussed what they had seen before they individually completed a recall test.
It was found that 71% of participants recalled aspects of the event that they had not seen for themselves, but had learnt during the discussion. This concludes that witnesses often go along with each other, either to win social approval or because they believe the other witnesses are right and they’re wrong. This is called memory conformity.

Evaluation:

P: the consequences of inaccurate memory recall are minimal in research settings compared with real life incidents.
E: Foster et al (1994) showed EWT was more accurate for real life crimes, opposed to research experiments
E: this means the findings of supporting research may lack validity and therefore shouldn’t be generalised to eye witness testimonies in real life.

Effects of anxiety:

Anxiety is…a state of emotional and physical arousal. It is a normal reaction to stress and some emotional side effects are having worried thoughts or feelings like tension. Physical side effects include increased heart rate or sweating.

Negative effects due to anxiety

Anxiety creates physiological arousal in the body which prevents us from paying attention to important cues, and so recall is worse.

Johnson and Scott (1976) investigated the effects of weapons: participants were led to believe they were taking part in a laboratory experiment and whilst waiting in the waiting room, the heard an argument in the next room.
In one condition, a man walked through the waiting area carrying a pen with grease on his hands.
In the second condition, a man walked out holding a blood covered paper knife.
Participants had to later pick out the man from a selection of 50 photographs.
49% who saw the pen were able to identify him, whist only 33% who saw the knife were able to.
This tunnel theory of memory suggests that a witnesses attention focuses on the weapon rather than the person.

Positive effects due to anxiety

The stress of witnessing a crime or accident creates anxiety through physiological arousal within the body. The fight or flight response is triggered which increases our alertness and improves our memory for the event because we become more aware of cues in the situations.

Yuille and Cutshall (1986)

They interviewed 13 witnesses to a real-life shooting – in which a Canadian store owner was injured and the thief was shot dead.
Some of the participants had been close to the incident whilst others had viewed it from further away.
These interviews were held 4-5 months after the incident and were compared with the original statements given at the time of the shooting.
The accuracy was determined by the number of details reported in each account.
They were also asked to rate how stressed they had felt at the time, using a 7-point scale asking about problems such as sleeplessness.
It was found that the witnesses accounts were accurate and little change had occurred over the 4-5 months. The participants who had reported high levels of stress were the most accurate (88% compared to 75% in the less stressed group)

Explaining the contracting findings

According to Yerkes-dodson, the relationship between emotional arousal and performance looks like an inverted U.
Deffenbacher (1983) applied this law to eye witness testimonies:

Lower levels of anxiety = lower accuracy of recall, and as anxiety increases so does recall accuracy, but – it reaches an optimal level of anxiety: if more anxiety than this is experienced, the accuracy decreases dramatically.

Evaluations:

P: weapon focus may be irrelevant
E: may test surprise rather than anxiety. Pickel (1998) showed a hair salon video with different hand-held items (scissors, gun, wallet or raw chicken) – accuracy was worse in the highly-unusual conditions (chicken and gun) – suggesting that weapon focus is due to unusualness rather than anxiety/threat.
E: explains nothing about anxiety’s effect of EWT and therefore low validity

P: real life witnesses are interviewed after events
E: things could happen to participants over this time ie. post-event discussion or media accounts of the incidents
E: these extraneous variables cannot be controlled and therefore may be the reason for accurate recall – potentially low validity

P: demand characteristics
E: most lab studies show filmed (mostly staged) crime and participants are likely to figure out they’re going to be questioned on what they observed
e: low validity and shouldn’t generalise to real life.