Bruce And Frowd: Facial Composites (Witnesses)

Question 1

Method?

Answer 1

3 lab experiments

Question 2

Measure?

Answer 2

Independent measures

Question 3

PPs?

Answer 3

EXP1: 30 staff+students from Stirling UNI
EXP2: 48 undergraduates at Stirling UNI
EXP3: 8 staff+students from computer science+psychology as Stirling UNI

Question 4

Procedure of EXP1?

Answer 4

• stimuli= target photos of 10 celebrities+50 composite images
• every face clean shaven+spectacles avoided
• 3 types of composites: ‘complete’ ‘internal’ ‘external’ (conditions)
• asked to place each composite in front of a celebrity face in own time

Question 5

Results of EXP1?

Answer 5

• ‘complete’+’external’= 35% correct

* internal= 19.5%

Question 6

Procedure EXP2?

Answer 6

• photo array with distracter faces (foils) making task harder
• foils+faces either made easy (different) or hard (similar)
• composites then presented 1 at a time along with photo array+pp had to pick the celebrity which matched composite
• composites either internal/external/complete

Question 7

Results EXP2:

Answer 7

• external= 42% correct
• internal= 24% correct

Both consistent across array type (hard or easy)

Question 8

Conclusions of EXP1+2?

Answer 8

• something about internal features of a face which don’t work well when reconstructing
• holds true even when the face is familiar

Question 9

Procedure of EXP3?

Answer 9

• familiar or unfamiliar target faces presented for 30s to witness who then created Pro-fit composite images
• ^these then photoshopped to create internal/external composites
• familiar/unfamiliar faces= lecturers/professors
• composites then randomly presented to ‘sorters’ who replicated EXP1
• sorters= 30 female+24 male

Question 10

Results EXP3?

Answer 10

• avg 57.7% accuracy
• composites of external features= 53.3%
• internal= 32.6%
• complete= 61.1%

Question 11

Strengths?

Answer 11

• concurrent validity
• independent measures= no order effects
• lab exp= control of extraneous variables

Question 12

Weaknesses?

Answer 12

• independent measures= individual differences

* low Eco validity= eye witness not as controlled, unlikely to be familiar

Question 13

Aim?

Answer 13

To investigate the relative recognisability of internal and external features of a facial composite.