The are definitively seen in neural correlates whenever

Effect Emotion has on the Perception of Race


The University of the South











recent shift of focus could be aiding in negative racial biases through media’s
use of persuasion. Recent research suggests that news media focuses a majority
of their time on racial discriminatory acts in society in both a negative and
neutral light (Schemer, 2012); thus society receives a majority of news
involving race. Due to the fact that the internet is consistently used for
receiving news reports (Zúñiga, Jung, & Valenzuela, 2012), a lot
of racial discrimination could be constantly streamed into the minds of every
individual in society; whether it be direct (via. News applications) or
indirect (via. Facebook; Tynes & Markoe, 2010), individuals are biased to
personal beliefs (e.g. political affiliated individuals will focus on things
pertaining to their political party’s beliefs; Krueger, 1996). Though the
population could receive any recent news occurring in the world, research
suggests that one of the media’s main focuses are minorities and discriminatory
acts involving these groups (Bleich, Stonebreaker, Nisar, & Abdelhamid,
2015). Therefore, it could be difficult to alter one’s own perception of race
from that of popular media opinion. Research also supports that context plays a
role in how race is perceived (Ito, Willadsen-Jensen, Kaye, & Park, 2011).
This implies that due to the negative context in which minorities are focused
on in media, increased negative bias may affect the perception of race outside
of media observation.

            Certain biases are
definitively seen in neural correlates whenever presented with stimuli
involving race. For example, the N170 amplitude is often correlated with facial
recognition, though recent research indicates that alongside this, racial
perception is noticed (Wiese, Kaufmann, & Scheinberger, 2012). Wiese and
colleagues (2012) found higher amplitudes in the N170 whenever presented with
faces of the same race, and a higher N250 amplitude whenever presented with
different races. This implies that own race is recognized quicker than opposing
races; however, those of a different race show higher perception upon
recognition. Further research supports this by showing higher P200 and N200
amplitudes which correlate with attention and opposition detection respectively
(Ito & Bartholow, 2009). Therefore, alongside behavioral support, more
definitive neural analysis demonstrates differentiation of racial perception
among the population.

can often arise due to the affect correlated with them (e.g. negative biases
frequently arise due to prior negative emotion; Segerstrom, 2001). Given the role that media plays, it
makes sense that emotional shifts can be due to the intake of information from
media outlets. In fact, social media is seen to promote certain attitudes about
subjects which further shift behaviors in situations involving these groups,
which consistently reinforces biases through negative wording (Zúñiga, Jung,
& Valenzuela, 2012). Specifically, the media aims to promote emotions that
seem elicit certain behaviors on particular groups of people (e.g. mass
shootings and race of the perpetrator). Schmid and Amodio (2016) conducted a
study on the manipulation of power and its effects on stereotyping and found
that increases in power are directly correlated with increased stereotyping,
suggesting that the media, due to its power, promotes an increase in racial
stereotyping. Further research specifically found correlation with negative
emotional wording in news media used to persuade viewers in a certain direction
(Dunlop, Wakefield, and Kashima, 2008), promoting media power over the
population. As previously mentioned, the media is a tool utilized by a majority of the population (Zúñiga, Jung,
& Valenzuela, 2012), suggesting that the population provides power to media
groups. In order to understand what is happening in the world, news outlets
stream consistently, and depending on personal viewing preference, certain
biases of these outlets shift the opinions of the viewers. Thus, media biases
arguably infect the biases of the population.

is a psychological tactic used to shift preference in a particular direction
(Tversky & Kahneman, 1981), and recent research shows news media utilizing
framing to alter viewer preference (Van Der Meer & Verhoeven, 2013). Racism occurs actively in society
(Kawakami, Dunn, Karmali, & Dovidio, 2009) and news media frames minority
groups (Bleich, Stonebreaker, Nisar, & Abdelhamid, 2015) negatively;
therefore, it is probable the majority of society assimilates to these negative
biases of the media. Kim’s (2011) study shows that people trust news coverage
without much question, even though media uses persuasion to elicit negative
emotions alongside minority coverage (Dunlop, Wakefield, & Kashima, 2008).
Further research suggests that media uses framing to elicit the responses they
intend (An & Gower, 2009). Framing is a technique in which words and
phrases are specifically chosen to shift the opinion of the listener to that of
the one using the framing (Tversky & Kahneman, 1981). Though some
research suggests that TV media plays an indirect role in emotional responses
to recent events (Namkoong, Fung, & Scheufele, 2012), other research
suggests that news coming from social media (e.g. Facebook) can play a direct
role in changing emotional responses to social situations (Zuniga, Jung, &
Valenzuela, 2012). For example, recent media framing of mass shootings
involving an African American perpetrator revolve around negative phrases and words;
whereas those involving a Caucasian perpetrator are often framed with positive
events in the criminal’s life. With so many different ways of getting media
information, information that is shown to direct coverage on minorities and
elicit negative emotions, it may be seen that negative biases on race can be
extorted whenever a negative emotion is felt regardless of context.

abundance of evidence supports the claim that media uses persuasion to elicit
negative biases centered on minority groups (Bleich, Stonebreaker, Nisar, &
Abdelhamid, 2015; Schemer, 2012; Dunlop, Wakefield, & Kashima, 2008). The
present study aims to provide additional support this by inducing negative and
positive moods in participants and showing images including race, and analyzing
the N170 and N250 components of ERPs in the parietal and temporal lobes
respectively. The components are seen to be directly involved in racial
processing (Senholzi & Ito, 2012; Wiese, Kaufmann, & Schweinberger,
2012; Ito & Bartholow, 2009; Ito & Urland, 2005), and will give more
specific data on how race plays an effect whenever emotion is induced. As a
result of prior evidence supporting negative framing correlation with negative
biases, it is expected that those induced in a negative emotion will inherently
rate non-racial images (i.e. images involving race but are not violent or
discriminatory) as less appropriate than those induced with a positive mood.
Alongside behavioral responses, it is expected that the component N250 will
show lower arousal in those induced into a negative mood, indicating that
negative affect causes a blockage of racial processing thus forcing
participants to rely on subjective biases to make judgement. It is also
expected that N170 amplitudes will be greater in every participant that views
their identified race.



            Our participants were 18 volunteers from Sewanee: the
University of the South. These participants were recruited from the Affective
Neuroscience course where the experiment took place. The participants consisted
of X males and X females between the ages of X
and X (mean age of X). The International Review Board
approved the study and all participants provided informed consent before
beginning the experimental task.


            The materials in the present study consisted of a
demographic questionnaire, the Positive and Negative Affect Schedule (PANAS),
the State-Trait Inventory for Cognitive and Somatic Anxiety (STICSA), the
Emotion Regulation Questionnaire (ERQ), the Brief Edinburgh Handedness
Inventory, a Racism Questionnaire, and the NASA Task Load Index (NADA-TLX). The
study also consisted of a computerized experimental portion that was coded and
analyzed using E-Prime and images and emotionally primed words obtained and
created by the researchers.

            Questionnaires. The
demographic questionnaire, PANAS, STICSA, ERQQ, and the Handedness Inventory
were administered before the experimental portion of the experiment. This was
to obtain baseline information about each participant’s emotional state and
their ability to regulate emotions. The demographic questionnaire consisted of
questions asking each participant’s age, sex, gender, ethnic background, level
of education, socioeconomic status, residence location (e.g. suburb, rural
areas, etc.), and their vision and if they wear contacts (this is due to the
EEG experiment and possible noise due to eye dryness and movement).


            The independent variable in this study is the emotional
stimuli, this includes either a positive emotion or a negative emotion before
being shown images involving race. The dependent variable of this study is the
participant’s reaction to the images involving race. ERP components N170 and
N250 will be examined from the parietal and temporal lobes to assess how
quickly each participant responds to the images in comparison to the responses
they give to each image.

Recording and Preprocessing

EEG recordings were made from Brain Products ActiCHamp system (Brain Vision,
LLC, Morrisville, NC) consisting of 64-electrodes arranged in an actiCAP
elastic cap and placed in accordance to the 10-20 System. The EEG signal was
sampled at a rate of 500 Hz and referenced to Cz. Impedance levels were kept
below 10k? at all sites. Offline, preprocessing was conducted using MATLAB
(version 2013b, The MathWorks, Inc., Natick, MA) along with the EEGLAB toolbox
(Delorme & Makeig, 2004). To derive the N170 and N250 waveforms, EEG data
were filtered from .1 to 30 Hz and re-referenced to the common average
reference. Single-trial EEG epochs were extracted for a period beginning 200ms
before stimulus onset and continuing for the entire duration of the racial
image presentation (1500 ms). Epochs were baseline-corrected using the 200 ms
prior to stimulus onset. Trials were discarded due to excessive physiological
noise if they contained: (i) an eye-blink, (ii) a voltage step greater than 50
mV/ms between sample points, (iii) a max-min difference greater than 150 mV/ms
throughout the epoch and (iv) low activity (i.e. less than 0.5 mV/ms) within a
100 ms window. Consistent with prior research, we quantified the N170 as the
average signal amplitude at site Fp1 and Fp2 in the 130-200 ms and the N250 at
site T7 and T8 in the 235-335 time range after stimulus onset.


            Upon entering the lab, participants provided informed
consent were then fitted with an electrode cap for electroencephalographic
(EEG) recordings. While the electrode cap was being set up, the participant
completed the demographic questionnaire, PANAS, STICSA, ERQ, and the Handedness
Inventory. After these were finished, the participant was lead into the
acquisition room where the EEG cap was placed on the participant and the
electrodes were gelled and tested. Once the EEG cap was finished and working
properly, the participant took a short practice experiment to fully understand
what was expected of them before the trials and recording began.

            One the practice experiment was completed, the
participant was able to ask any questions or concerns they may have. During the
experimental task, the participant was primed with either a positive (joy) or
negative (angry) word before being presented with a racial image including a
blatantly racist scene or a scene that did not include racism and was tasked
with rating each image on a scale from 1 (not acceptable) to 5 (extremely acceptable).
The participant was assured that their responses were completely confidential
and that responses were kept between the researchers only.  Each block contained 100 trials consisting of
50 blatantly racist images and 50 non-racism images preceded by 75
positive/negative words and 25 positive/negative words depending on the block’s
emotion induction. The reason for the emotional stimuli being 75:25 is so the
participant does not figure out what the hypothesis is during the experimental
portion of the task.

            Upon completion of the experimental portion, the
participant was administered a racism questionnaire, to assess their baseline
racial opinions, and the NASA TLX, to assess how stressful they thought the
task to be. The reasoning for the NASA TLX is to test whether or not cognitive
load depletion may be a factor when perceiving the racial images.


            Behavioral results expected will show a quicker reaction
time for those induced in a negative mood. To test this, a T-test would be
performed comparing each participant’s average response time for both positive
and negative conditions. To compare perception of racism and mood, another
T-test will be performed analyzing each participant’s response to each image
against the mood induced in that block. It is expected that whenever induced
into a negative mood, higher unacceptability will be recorded in the responses.
Overall, negative mood will show each image to be less acceptable; however,
comparing positive and negative moods for the non-racial images, negative mood
induction will positively correlate with a lower acceptance rating.

            Based on previous research in the N170 and N250
components in ERPs, it is expected that higher N170 amplitude will be recorded
whenever each participant examines an image involving their own race. In these
same participants, a higher N250 will be recorded when images show an opposing
race. The N250 will also show higher amplitude in all instances involving
induction into a negative mood for both racially explicit and non-racial
images. In recording the P300, lower amplitude will be seen when induced in a
negative mood for both racially explicit and non-racial images. Although,
whenever induced in a positive mood, the P300 will show higher amplitude in
only the racially explicit images.

            ERP results will look similarly to the epochs in figure
one. The non-racist images epoch shows a greater amplitude for N170 when
looking at the negative emotion, suggesting that negative affect increases
response to faces more readily than those in positive moods. This epoch also
shows a greater N250 amplitude for those induced into a positive mood,
suggesting that negative affect does not recognize race as much as positive
affect. As hypothesized, there should be no significant difference in N170
amplitude for racially-explicit images between moods, though negative emotion
should have lower amplitude for the N250; this is shown under the racist images
epoch. The remaining two epochs show definitive differences between affect and
perception of race, controlling for the overarching assumption that mood
affects racial perception.


            Given that the expected results hold true, definitive
implications can be made from both the behavioral and EEG data about how mood
affects the perception of social issues. Induction into a negative mood will
cause a quicker response to acceptability for both non-racial and racially
explicit images. This would indicate that whenever negative affect is
inherently present, focus on surroundings lessens; thus, one would rely on
unspoken biases in reaction rather than acting reasonably. These biases may
stem from media intake and not be core opinions that people hold. This suggests
that negative affect takes up higher amounts of cognitive load and therefore
causes unacceptable societal actions.

            With the indicated results, lower amplitudes in the N250
in negative mood induced individuals further supports that negative affect
causes an increased cognitive load and enforces reaction from subconscious
biases. This suggests that, because of the biases media produces, violent
behavior may arise, not due to negative intention, but due to the inability to
regulate actions because of lower cognitive abilities during emotionally taxing
situations. Though this in no way justifies negative behavior towards
minorities, it does give further explanation as to why many individuals act
irrationally in certain situations when they may not usually react in such a


            Any limitations for conducting this study would be
centered on the population used. For example, college students may show less
unconscious bias than older generations, or college students could show greater
bias due to their amount of media intake. It will also be difficult to conclude
that unconscious biases are solely due to lower cognitive abilities to
sufficiently regulate actions against minorities. There are numerous factors to
take into consideration whenever attempting to pinpoint reasonings for
discriminatory behaviors. Though, any research in this area is beneficial for
lessening irrational biases learned from media.


            These findings support Windmann and Kr?gert’s (1998)
study suggesting unconscious information processing leading to
misinterpretations. Since the presenting study suggests that those induced into
a negative misconstrue non-racial images as more unacceptable, it can be
hypothesized that subconscious biases play a role in reactions to certain
situations. This data also slightly supports the unconscious bias theory. This
theory was originally used to further understand discrimination in the
workplace (Lee, 2005); however is touches on the idea that discrimination does
not, necessarily, stem from core beliefs, but rather from past stereotypes
learned either from previous generations or simply from the learning of
history. Though there are numerous theories behind discrimination, it is still
unknown directly from which they stem. Thus, more research is needed to fully
identify and define this theory of unconscious biases that way preventions
could be taken to eliminate unhealthy and irrational biases that cause violent
behavior in society.













Bleich, E., Stonebraker, H., Nisar, H., &
Abdelhamid, R. (2015). Media Portrayals of Minorities: Muslims in British
Newspaper Headlines, 2001–2012. Journal of
Ethnic and Migration Studies,41(6), 942-962. doi:10.1080/1369183x.2014.1002200

Ito, T. A., & Urland, G. R. (2005). The
influence of processing objectives on the perception of faces: An ERP study of
race and gender perception. Cognitive,
Affective, & Behavioral Neuroscience,5(1), 21-36. doi:10.3758/cabn.5.1.21

Ito, T. A., & Bartholow, B. D. (2009). The
neural correlates of race. Trends in
Cognitive Sciences,13(12),
524-531. doi:10.1016/j.tics.2009.10.002

Kim, K. S. (2010). Public understanding of the
politics of global warming in the news media: the hostile media approach. Public Understanding of Science,20(5), 690-705. doi:10.1177/0963662510372313

Krueger, J. (1996). Personal beliefs and
cultural stereotypes about racial characteristics. Journal of Personality and Social Psychology,71(3), 536-548. doi:10.1037/0022-3514.71.3.536

Lee, A. J. (2005). Unconscious bias theory in
employment discrimination litigation.

Meer, T. G., & Verhoeven, P. (2013). Public
framing organizational crisis situations: Social media versus news media. Public Relations Review,39(3), 229-231. doi:10.1016/j.pubrev.2012.12.001

Namkoong, K., Fung, T. K., & Scheufele, D.
A. (2012). The Politics of Emotion: News Media Attention, Emotional Responses,
and Participation During the 2004 U.S. Presidential Election. Mass Communication and Society,15(1), 25-45. doi:10.1080/15205436.2011.563894

Schemer, C. (2012). The Influence of News Media
on Stereotypic Attitudes Toward Immigrants in a Political Campaign. Journal of Communication,62(5), 739-757. doi:10.1111/j.1460-2466.2012.01672.x

Segerstrom, S. C. (2001). Optimism and
Attentional Bias for Negative and Positive Stimuli. Personality and Social Psychology Bulletin,27(10), 1334-1343. doi:10.1177/01461672012710009

Tversky, A., & Kahneman, D. (1985). The
Framing of Decisions and the Psychology of Choice. Behavioral Decision Making,25-41. doi:10.1007/978-1-4613-2391-4_2

Tynes, B. M., & Markoe, S. L. (2010). The
role of color-blind racial attitudes in reactions to racial discrimination on
social network sites. Journal of Diversity in
Higher Education,3(1), 1-13.

Windmann, S., & Krüger, T. (1998).
Subconscious Detection of Threat as Reflected by an Enhanced Response
Bias. Consciousness and Cognition,7(4), 603-633. doi:10.1006/ccog.1998.0337

Zúñiga, H. G., Jung, N., & Valenzuela, S.
(2012). Social Media Use for News and Individuals Social Capital, Civic
Engagement and Political Participation. Journal
of Computer-Mediated Communication,17(3), 319-336. doi:10.1111/j.1083-6101.2012.01574.x