Abstract

1
The Development of the Own-Race Bias from 7 years
to Adulthood in Typical Development and
Autism Noah J. Sasson and J. Steven
Reznick University of North Carolina at Chapel
Hill
Abstract Individuals exhibit a marked advantage
for the discrimination and recognition of
well-experienced same-race faces relative to
less-familiar other-race faces, a phenomenon
often referred to as the own-race bias (ORB). If
differential facial experience throughout
development contributes to the establishment of
the ORB, recognition memory should improve with
age to a greater degree for same-race faces
compared to other-race faces. 72
typically-developing (TD) Caucasian
seven-year-olds, 10-year-olds, and adults viewed
48 faces (24 Caucasian, 24 Korean) one at a time
for 2 s. each. 48 more faces were then displayed
(half of which were shown earlier) and
participants judged whether a face was previously
presented. A counterbalanced inverted condition
was also included. A three way interaction
between age group, orientation and stimulus race
emerged, indicating that the ORB increased with
age, but only in the upright condition. A pilot
sample of 15 children and adults with
High-Functioning Autism (HFA) was also tested in
order to explore whether the developmental
pattern differs for this population.

• Stimuli
• 72 male Korean faces and 72 male Caucasian faces,
  each approximately 5.5 x 8.0 cm in size, were
  used. All displayed a neutral expression, were
  free of extraneous accessories such as eyewear,
  and were cropped in order to ensure processing of
  only the internal section of the face.

• Results, cont.
• Performance was better on upright faces than
  inverted faces, F(1, 69) 69.20, p lt .01, and on
  Caucasian faces as compared to Asian faces, F(1,
  69) 81.81, p lt .01.
• There was a significant interaction between
  orientation and stimulus race. F(1, 69) 50.66,
  p lt .01.
• There was an overall effect of age, F(2, 69)
  24.67, p lt .01. Adults gt 10-year-olds gt
  7-year-olds.
• Significant two-way interactions emerged for Age
  X Orientation, F(2, 69) 14.60, p lt .01, and Age
  X Stimulus race, F(2, 69) 3.77, p lt .05.
• A significant three-way interaction emerged
  between Age X Orientation X Stimulus race, F(2,
  69) 4.83, p lt .05. That is, the ORB increased
  from 7 years to adulthood for upright faces. The
  youngest participants still exhibited greater
  recognition for upright Caucasian faces than
  upright Asian faces, t(23) 3.158, p lt .01
• Exploratory data analysis on participants with
  HFA revealed that they appear to perform less
  well than their TD counterparts on upright
  same-race faces, but at least as well on upright
  other-race faces and both sets of inverted faces.

Figure 1. Examples of same and other race face
stimuli used in the ORB task.

• Procedure
• Encoding phase 48 faces (24 Caucasian, 24
  Korean) shown for 2 s. each with an interstimulus
  interval of 1 s.
• Recognition Phase 48 more faces shown (half
  previously displayed) one at a time until
  participant responds via the keyboard whether the
  face was seen before.
• Also done in a counterbalanced inverted condition.

• Introduction
• The own-race bias (ORB) refers to the robust
  finding that individuals demonstrate a memory
  advantage for faces belonging to their own race
  compared to faces of another, less familiar race
  (Chance and Goldstein, 1996 Meissner Bringham,
  2001)
• There are conflicting findings concerning when
  the ORB emerges and stabilizes (Chance, Turner
  Goldstein, 1982 Pezdek, Blandon-Gitlin Moore,
  2003 Sangrigoli de Schonen, 2004)

• Discussion
• These findings suggest that the ORB is present by
  7 years of age and continues to increase until
  adulthood. This age-related increase in the ORB
  may reflect the development of facial expertise
  strategies for well-experienced same-race faces.
• Developmental changes in memory cannot account
  for the age-related changes because recognition
  memory improved with age to a greater degree for
  same race faces than other race faces.
• Inversion eliminates the age-related increase of
  the ORB. Same-race faces show a larger inversion
  effect than other-race faces, and this
  discrepancy increases with age.
• Individuals with HFA may exhibit less of an ORB
  than TD participants

Results

• Overall performance for TD participants

• Scatterplots for participants with HFA (capped at
  age 22) relative to TD mean

Upright Total Inverted Total
Upright Caucasian Upright Asian

• Methods
• Participants
• 72 TD individuals split evenly into three age
  groups 7-year-olds (M 7.10, SD .44),
  10-year-olds (M 10.01, SD .63), and adults (M
  19.72, SD 1.44).
• Pilot sample of 15 individuals with HFA split
  evenly into three age groups 7 to 10 (M 8.51,
  SD 1.05), 11 to 13 (M 12.52, SD .82), and
  adults (M 34.53, SD 7.21).
• All participants screened for their relative
  experience with Caucasian and Asian faces.

Inverted
Caucasian Inverted Asian
Acknowledgments This research was supported by a
National Science Foundation Graduate Research
Fellowship.
Figure 3. Percentage correct for individual
participants with HFA relative to TD mean
Figure 2. Percentage correct for each age group
on each condition