eyewitness testimony (EWT)

+ Gabbert et al.

describe response bias explanation:

• suggests that Q wording has no effect on the memory of ppts - instead influences how they decide to answer

• e.g. Loftus and Palmer (1974) “smashed” encouraged ppts to give a higher speed estimate than they remembered

describe substitution explanation:

• suggests Q wording changes ppts’ memory

• Loftus and Palmer’s second experiment - ppts who heard “smashed” were more likely to report seeing broken glass

• → critical verb altered memory

summarise Clifasefi et al. (2013):

• used food/drink profiles to implant a false memory into ppts when asked to recall whether or not they had ever been drunk to the extent that they were sick

• many ppts recalled being sick from drinking, w/ some claiming they disliked certain alcoholic drinks because of it

• demonstrates substitution explanation - misleading info altered memory of ppts as their previous memory had been overwritten

what does the Yerkes-Dodson Law state?

moderate levels of arousal (stress) → optimum performance

summarise Johnson and Scott (1976):

• investigated weapon focus - volunteers in a lab setting witness a row in either low anxiety (greasy pen) or high anxiety (bloody knife)

• → low anxiety associated w/ high accuracy of EWT: anxiety has a -ve effect on recall

give a +ve of Johnson and Scott (1976):

lab experiment - high intval due to good variable control

give 2 -ves of Johnson and Scott (1976):

• no protection from psychological harm - induced anxiety

• may test surprise rather than anxiety

summarise Yuille and Cutshall (1986):

• witnesses to a real life shooting in which 2 men were shot, 1 fatally, were interviewed 4-5 months after the incident

• account compared to original account

• high anxiety associated w high accuracy of EWT: anxiety has a +ve effect on recall

give a +ve of Yuille and Cutshall (1986):

high ecoval - ppts experienced real anxiety in a real setting

give a -ve of Yuille and Cutshall (1986):

not all witnesses agreed to be reinterviewed - so may not be a representative sample (e..g those most traumatised may have been more likely to decline a second interview)

summarise Parker et al. (2006):

• interviewed people affected by a hurricane to find a relationship between memory of events and anxiety (operationalised by damage to homes)

• → moderate levels of anxiety associated w/ high EWT accuracy: anxiety can have a +ve or -ve effect on recall depending on its extremity

give 2 +ves of Parker et al. (2006):

• high ecoval - ppts experienced real anxiety in a real setting

• investigated moderate levels of anxiety (as well as high and low), allowing for a better understanding of the relationship between anxiety and witness accuracy

give a -ve of Parker et al. (2006):

operationalisation of anxiety through damage may not reflect experienced anxiety

summarise Valentine and Mesout (2009):

• visitors to a horror labyrinth /ed into low and high anxiety on basis of heart monitor

• asked to describe an individual encountered in labyrinth

• → low anxiety associated w/ high EWT accuracy: anxiety has a -ve effect on recall

give 2 +ves of Valentine and Mesout (2009):

• high ecoval - real life setting

• heart monitor accurate measure of anxiety

give 2 -ves of Valentine and Mesout (2009):

• not so high ecoval after all - anxiety not caused by anything really threatening

• Quasi experiment - no random allocation to conditions, ppt variables may have acted as confounding variables

name and explain 2 factors affecting identification:

• misleading info e.g. leading Qs/PED

• anxiety e.g. weapon focus having +ve/-ve effects

describe Loftus and Palmer’s initial (1974) experiment:

• aimed to see if people could be lead to remember a scene that never happened

• 45 students shown 7 clips of road traffic accidents

• all asked “how fast were the cars going when they X?”

• X substituted for contacted/hit/bumped/collided/smashed

• “contacted” resulted in lowest speed guess, “smashed” highest, w/ a difference of almost 10 mph

• then asked “ did you see any broken glass was seen when X?” “hit” - 4% said yes, “smashed” - 32% said yes (there was no broken glass)

give 3 strengths of Loftus and Palmer (1974):

• tempval and reliable - repeats

• lab study - high variable control, replicable

• RWA - courts

give 4 limitations of Loftus and Palmer (1974):

• lacks ecoval

• individual differences

• lacks popval

• order effects

describe Loftus and Palmer’s additional study:

• 2 groups - 1 control, 1 experimental

• control “did you see a broken headlight?”

• experimental “did you see the broken headlight?”

• 2x as many people in the experimental group said they did (there was no broken headlight)