Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Biological Basis of Behaviour II > Behavioural Facial Recognition > Flashcards

Behavioural Facial Recognition Flashcards

1
Q

FACIAL RECOGNITION

A
  • one of the best cognitive skills we have
  • can rapidly detect face in visual scene; can recognise familiar face in milliseconds
  • can extract relevant info necessary to categorise person’s facial expression/gender/race/gaze direction
  • accurate face recognition = crucial in social interaction; gives clues about who we interact w/ & their motivational/emotional states
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

SPECIFICITY ACCOUNT

A

MCKONE & KANWISHER (2005)
- neuro-cognitive mechanisms = selectively involved in processing faces “per se” w/little role in processing non-facial stimuli

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

EXPERTISE ACCOUNT

A

GAUTHIER & TARR (1997)
- neuro-cognitive mechanisms involved in face processing = elicited for all prototype-defined stimuli (aka. have shared configuration) as long we we’ve been pre-exposed

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

THE FACE INVERSION EFFECT (FIE)

A

YIN (1969)
- decrement in recognising/discriminating between inverted/upright faces performance
- long-standing debate over if FIE due to faces = special; subject to analysis/representation by dedicated neural mechanisms specific to faces OR faces = expertise effect aka. lots of experience w/upright faces; allows expertise build-up
- results in better recognition of upright faces
- expertise does NOT apply to inverted faces (aka. performance dif)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

FIE: SPECIFICITY

A

YIN (1969)
- forced choice old/new recognition task
- pps shown individual pictures (study phase); presented w/picture pairs (recognition task)
- recognition task = indicated pair they thought they’d seen in study phase
- exp = orientation of materials is same (upright/inverted) in study phase/recognition task

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

FIE: SPECIFICITY (MATERIALS)

A
  • 64 dif pictures (black/white)
  • each of 4 material types (faces/houses/airplanes)
  • all pasted on card
  • studio picture faces of adult males; similar age/expression/lack of outstanding distinguishing features (ie. glasses/beards/unique marks
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

FIE: SPECIFICITY (PROCEDURE)

A
  • each pp looks at 40 single photos (3sec per photo)
  • recognition task = 24 picture pairs (16 study phase not shown); each contained 1 old (exact study phase duplicate) & new picture (not previously shown); pp must indicate old photo (at own rate)
  • all pps went through 4 study phase/recognition task blocks (2 upright; 2 inverted); half pps saw upright first; half inverted first
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

FIE: SPECIFICITY (RESULTS)

A
  • t-tests for matched pairs (2-tailed)
  • inversion effect = greatest for faces (t(25) = 8.48, p < .001(STATSIG))
  • BUT weaker in houses (t(25) = 3.01, p < .01) & men in motion (t(25) = 2.15, p < .05)
  • NOT statsig in airplanes (t(25) < 1)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

FIE: EXPERTISE

A

DIAMOND & CAREY (1986)
- challenged idea of faces being “special”; introduced “expertise” as contributing factor to inversion effect
- demonstrated that large inversion effect (usually recorded for faces) can be obtained for dog images when pps = dog breeders (ie. experts)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

FIE: EXPERTISE (MATERIALS)

A
  • photos of dogs/human faces on index cards
  • whole-body/3 quarter profile champion dog photos selected from NYC American Kennel Club (AKC) archives
  • set of photos (1/2 men; 1/2 women) taken from college yearbooks; photos cropped beneath chin to eliminate clothing; no distinguishing features
  • novice pps = undergraduates
  • expert pps (16) = dog experts from AKC Directory of Dog Show & Obedience Judges
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

FIE: EXPERTISE (RESULTS)

A
  • predicted 3-way interaction of subject group/material/orientation = STATSIG (F(2, 45) = 9.40, p < .001)
  • experts showed similar inversion effect for faces/dogs
  • showed it’s possible to obtain robust inversion effect for dog images AND faces IF pps = specific dog breed experts
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

FIE: MAIN FINDINGS

A

YIN (1969)
- robust inversion effect for faces larger than for other stimuli sets supporting specificity account of face recognition mechanisms
DIAMOND & CAREY (1986)
- robust inversion effect for dog images like faces (when pps = expert dog breeders) supporting face recognition expertise account

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

CONFIGURAL PROCESSING

A
  • to recognise faces we rely on experience w/configural info
  • configural processing = 1st (ie. spatial relations among main stimulus features)/2nd order (ie. variations in 1st order relations relative to prototype for stimulus set) relations sensitivity
  • also includes holistic processing (aka. stimulus = gestalt) so configural processing = spatial relations perception in stimulus
  • contrasts w/featural processing (perceiving specific features in isolation)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

FIE: CONFIGURAL PROCESSING

A

DIAMOND & CAREY
- expertise account
- inversion -> configural processing disruption -> reduced recognition performance
- robust inversion effect should be obtained for all stimuli sets sharing base configuration that we developed necessary expertise for

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

THE THATCHER ILLUSION

A

THOMPSON (1980)
- configural distortions’ sensitivity to inversion = oft suggest as base for “thatcherisation”
- illusion depends on mouth/eyes inversion being hard to detect in inverted face
- explanation = inversion reduces configural info in face; promotes more componential analysis of present features
- mouth/eyes = normal alone BUT + face -> configural processing -> distortions stand out -> strong pp reaction

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

CIVILE, COOKE ET AL. (2020): MATERIALS

A
  • 32 Exeter uni students = pps
  • 320 facial images (1/2 female; 1/2 male); neutral expressions
  • all faces standardised using greyscale colour on black background w/Gimp 2.6
  • hair/ears cropped; all faces Thatcherised & normally inverted (4 conditions)
17
Q

CIVILE, COOKE ET AL. (2020): PROCEDURE

A
  • old/new recognition task; 2 parts (study phase/old VS new recognition phase)
  • study phase = each pp shown 4 dif face type w/40 faces for each type
  • test phase = 160 novel stimuli of same 4 types; each never stimulated +1 condition
18
Q

CIVILE, COOKE ET AL. (2020): TRIAL STRUCTURE

A
  • pps looked at 160 facial images presented alone randomly; saw fixation cross in screen center presented for 500ms -> blank screen (500ms) -> facial stimulus (3000ms)
  • repeats until all stimuli done
  • old/new task = subjects shown familiar faces w/novels randomly
  • stimuli presented (4000ms); pps indicated recognition via (counterbalanced) keypress
19
Q

CIVILE, COOKE ET AL. (2020): RESULTS

A
  • 2x2 within-subjects ANOVA using face orientation (upright/inverted) & face type (normal/Thatcherised)
  • STATSIG interaction between face/type/orientation (p. = .009)
  • follow-up paired t-tests conducted to compare performance on each face type
  • large inversion effect for normal faces (p < .001)
  • reduced but still STATSIG inversion effect for Thatchers (p = .021)
  • STATSIG interaction interpreted as due to reduced inversion effect in Thatchers
  • upright normal recognition performance = STATSIG higher > Thatchers (p < .001)
  • no statsig dif found between inverted normal/inverted Thatchers
20
Q

THATCHERISATION: MAIN FINDINGS

A

DIAMOND & CAREY (1986)
- suggested expertise account where configural info perceptual processing = determinant of facial recognition
CIVILE, COOKE ET AL. (2020)
- demonstrated how configural info disruption via Thatcherisation significantly reduces FIE

21
Q

SUMMARY

A
  • literature evidence to support both facial recognition accounts: specificity VS expertise
  • 1 of the most replicated/robust cognitive phenomenon; FIE oft used to study face recognition skills nature
  • specificity = facial recognition is special as FIE = faces > objects
  • expertise = possible to get robust inversion effect for all stimuli sharing familiar configuration

Biological Basis of Behaviour II (25 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions