University of Puget SoundSound Ideas

Summer Research

Summer 2014

EEG study of perceptual bias in facial expressions,mood, and the mirror-neuron systemLaurel Olfsonlolfson@pugetsound.edu

Follow this and additional works at: http://soundideas.pugetsound.edu/summer_research

This Article is brought to you for free and open access by Sound Ideas. It has been accepted for inclusion in Summer Research by an authorizedadministrator of Sound Ideas. For more information, please contact soundideas@pugetsound.edu.

Recommended CitationOlfson, Laurel, "EEG study of perceptual bias in facial expressions, mood, and the mirror-neuron system" (2014). Summer Research.Paper 222.http://soundideas.pugetsound.edu/summer_research/222

Introduction • Recent studies suggest the mood-congruency perceptual biases,

that mood may alter our perceptions to interpret the world in a manner consistent with our mood (e.g. Lopez-Duran, Kuhlman, George, & Kovacs, 2013; Qin, 2012)

• Individuals in a somewhat depressed mood (but not clinically depressed) rated happy and neutral faces as having a lower intensity compared with individuals not in a happier mood (Qin, 2012)

• One explanation for the interaction between mood and perception of facial expressions may be empathy, or the ability to experience the emotions and feelings of another.

• The human mirror neuron system (MNS) may be an important brain system underlying social cognition abilities such as empathy (e.g. Carr, Iacoboni, Dubeau, Mazziotta, & Lenzi, 2003; Montgomery & Haxby, 2008)

• Prior research has shown the increased MNS activity is correlated with suppression of a particular waveform called the mu-wave. Thus, we measured mu-wave suppression as an indicator of MNS activity

• The potential link between the human MNS and differential perception of facial expressions given normal variations in mood is unknown

Stimuli

Questions • Can we replicate mood-congruency biases?

• Will perceptual biases relate to MNS activity?

• Will mu-wave suppression be greatest when viewing a mood-congruent face?

EEG • 32-channel Biosemi System was

used to record changes in electrical fields on the scalp due to brain activity

• Referenced to left and right mastoids

• Eye movements detected using electrodes around left eye

Acknowledgments This research was funded by a AHSS Student Summer Research Grant from the University of Puget Sound

Results

EEG study of perceptual bias in facial expressions, mood, and the mirror-neuron system

Laurel Olfson & David R. Andresen Department of Psychology, Neuroscience

References

Experimental Setup

• Data processed to remove eye movement and EMG artifacts using principle components analyses in EMSE EEG analysis software. Other artifacts removed using a recursive outlier detection algorithm in EMSE

• Electrodes were averaged across the entire brain

• EEG analyzed using fast-Fourier transform to produce spectral density plot

• Mu-wave power assessed as are under the curve between 8-13 Hz (integral)

• Mu-wave power normed by each subject’s overall average mu-wave power, such that 1.0 is average, >1.0 is increased mu-wave power and <1.0 is suppressed mu-wave power.

EEG Analysis Mood Assessment

Perceptual Mood Effects • Happy and sad faces were morphed using Fantamorph software

• The neutral, or ambiguous, faces were chosen by a selection of individuals not participating in the study. This face was not always the halfway point in the morph sequence. Task

Happy-Easy Happy-Hard Sad-Easy Sad-Hard Ambiguous

• Participants filled out several questionnaires assessing their current mood

• Participants then viewed faces via a computer monitor and indicated by pressing the keyboard whether they believed the face to be happy or sad while brain activity was measured via EEG

• Daily Life Questionnaire • (Schwartz, Romans, Meiyappan, DeSouza,

& Einstein, 2012)

• Mood Questionnaire • (Förster, Özelsel, & Epstude, 2010)

• Positive and Negative Aspect Scale • (Watson, Clark, & Tellegen, 1998)

• Analysis: Scores from the three questionnaires were summed. Higher scores indicated a happier mood, lower scores a sadder mood.

• A median split was used to create the ‘Positive’ and ‘Negative’ mood groups.

E.g. In the past day, how happy have you felt? 1 2 3 4 5 Not at all Very much

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

-100 -10 -5 -3 0 3 5 10 100

Pe

rce

pt

Re

spo

nse

s G

ive

n

Face Morph

Percent Responding “Happy” or “Sad”

Happy

Sad

Responses to Morphs EEG

The point of most ambiguity was found to be slightly ‘happier’ than what was deemed to be the midpoint (responded “Happy” and “Sad” equally often).

• Individuals with a more positive mood had more mu-wave suppression for happier faces than sadder faces.

• Those in a more negative mood had the opposite results: more mu-wave suppression when viewing sadder faces compared to happier faces.

• Suggests more mu-wave suppression when stimuli match mood.

Conclusions

Responded “Happy” Responded “Sad”

Happy-Hard Sad-Easy Sad-Hard Happy-Easy

Participants in negative mood had suppressed mu-waves when viewing sad faces compared to happy faces.

Participants in positive mood had suppressed mu-waves when viewing happy faces compared to sad faces.

Subjects • Results include 8 student subjects from the

University of Puget Sound.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

-100 -10 -5 -3 0 3 5 10 100

Pe

rce

nt

Re

spo

nd

ing

“Hap

py

”

Face Morph

Positive Mood Negative Mood

Happy-Hard Sad-Easy Sad-Hard Happy-Easy

• The point along the Happy-Sad morph continuum where participants responded equally as looking “Sad” or “Happy” differs depending on mood

• Being in a negative mood shifts the curve to the left from those in a positive mood , thus more of the morphed faces were perceived as “Sad” including those well into the “Happy” morphs.

• Morph manipulation accurately captured the transition between seeing a face as sad or happy

• Being in a negative mood shifts perception of ambiguous facial expressions towards looking more sad, relative to being in a positive mood

• EEG results show mu-wave suppression during mood-congruency conditions:

• Reduced mu-wave activity to happy faces for participants in more positive mood

• Reduced mu-wave activity to sad faces for participants in more negative mood

• Taken together, results suggest the human mirror neuron system is involved in mood-congruency perceptual effects

• Carr, L., Iacoboni, M., Dubeau, M., Mazziotta, J.C., & Lenzi, G.L. (2003). Neural mechanisms of empathy in humans: A relay from neural systems for imitation to limbic areas. PNAS, 100(9), 5497-5502. doi: 10.1073/pnas.0935845100

• Lopez-Duran, N.L., Kuhlman, K.R., George, C., & Kovacs, M. (2013). Facial emotion expression recognition by children at familial risk for depression: High risk boys are oversensitive to sadness. Journal of Child Psychology and Psychiatry, 54(5), 656-574. doi: 10.1111/jcpp.12005

• Montgomery, K.J. and Haxby, J.V. (2008). Mirror neuron system differentially activated by facial expressions and social hand gestures: A functional magnetic resonance imaging study. Journal of Cognitive Neuroscience, 20(10), 1866-1877. doi: 10.1162/jocn.2008.20127

• Qin, D. (2012). More excited for negative facial expression in depression: Evidence from an event-related potential study. Clinical Neurophysiology, 123(11), 2172-2179. doi: 10.1016/j.clinph.2012.04.018

0.97

0.98

0.99

1

1.01

1.02

1.03

Happy-Easy Happy-Hard Ambiguous Sad-Hard Sad-Easy

No

rme

d E

EG R

esp

on

se

Face Morph

Congruency between stimuli and participant mood causes increased mu-wave suppression

Positive Mood

Negative Mood