crowd emotion perception is ... - [email protected]
TRANSCRIPT
L R
pSTS
InsulaOFC
dmPFC
SMAdACC
RSC
PCC
V2
Research objectives • Understanding the factors that modulate crowd emotion perception and examining individual differences in this process
• Investigating neural bases for extracting crowd emotion and single face perception.
Different visual pathways: Magnocellular (M) and Parvocellular (P)
• Crowd emotion provides important social information guiding our interactions with others (e.g., approach to make friends or avoid them to be safe?).
• We can extract crowd emotion from multiple faces with different facial expressions very rapidly [1].
• However, little is known how this efficient, and socially important process is achieved.
• Distinct neural mechanisms seem to support crowd emotion perception and single emotional face perception: orbital, premotor, and frontal regions for crowd emotion perception and temporal areas for a single emotional face.
• One possibility: different contribution of Magno pathway (quick and dirty processing of a gist, relying on low spatial frequency information) for crowd emotion perception and of Parvo pathway for a single emotional face perception.
• Anxiety modulates crowd emotion perception: high anxiety individuals react to emotional crowds faster and show tendency to avoid happy crowds.
• Both behavioral and fMRI results show task goal dependent lateralization for crowd emotion, but not for single face perception.
• High anxiety individuals made faster responses and showed a tendency to avoid happy crowds.
• While a single angry face preferentially activated the cortical face network, a crowd of angry faces activated orbital, premotor and frontal regions (dorsal stream) as well as cerebellum and brain stem (PAG).
• Angry stimuli preferentially activated the brain regions for clear threat (Also check out the neighboring poster #63.4047), whereas happy stimuli preferentially activated the cortical face network.
Crowd
+5
-5
Dorsal pathway (Magno)
Ventral pathway (Parvo)
Magno pathway: more sensitive to low spatial frequency (e.g., global feature) Parvo pathway: more sensitive to high spatial frequency
Stimuli
Parvo cellMagno cell
And high-anxiety observers made faster responses than low-anxiety observers overall.45
55
65
75
85
5 15 25 35 45 55 65 750
0.5
1
1.5
2
5 15 25 35 45 55 65 750
0.5
1
1.5
2
5 15 25 35 45 55 65 75
RT
(sec
)
N = 90 r: -0.39, p < .001
Overall
STAI-T
STAI-T
45
55
65
75
85
5 15 25 35 45 55 65 7545
55
65
75
85
5 15 25 35 45 55 65 75
45
55
65
75
85
5 15 25 35 45 55 65 75
Angry crowd
Acc
urac
y (%
)
45
55
65
75
85
5 15 25 35 45 55 65 7545
55
65
75
85
5 15 25 35 45 55 65 7545
55
65
75
85
5 15 25 35 45 55 65 75
Happy crowd
STAI-T
N = 90 r: -0.46, p < .001
N = 90 r: -0.13, n.s.
• Visual field of presentation: Left, Right visual fields
• Anxiety: STAI [3]
• fMRI: 3T, TR:2s, 28 subjects (Avoidance only) Customized, manual shimming
“Which group (left or right) would you rather Avoid? Approach?
+Ready!
+
Stimulus (1s)
+
Blank (1.5s)
Crowd emotion perception is lateralized in a goal-driven fashion and modulated by observer anxiety and stimulus characteristics: behavioral and fMRI results
1Department of Radiology, Harvard University, 2Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, 3Department of Psychology, The Pennsylvania State University
Hee Yeon Im1,2, Daniel N. Albohn3, Troy G. Steiner3, Reginald B. Adams, Jr.3, Kestutis Kveraga1,2
#63.4048Background Summary
References
Funding: NIH R01MH101194 to K.K. and R.B.A., Jr.
Behavioral Results
Conclusion
Method• 50 morphed faces between Happy and Angry
…… 6 sets from different identities [2]
Single face trials
www.#kveragalab.org#
Angry Crowd vs. Single angry facefMRI Results
• Task goal-dependent laterality for crowd emotion Responses were more accurate when task-relevant crowd emotion was presented in left visual field (LVF).
Avoidance task (N = 21) Approach task (N = 21)
Acc
urac
y (%
)
60
75
70
65
Angry Happy Angry Happy Angry Happy Angry Happy
LVF RVF LVF RVF
But not for single faces!
Acc
urac
y (%
)
60
75
70
65
Angry Happy Angry Happy
LVF RVF
Avoidance task (N = 21)
• The effect of observers’ anxiety level
High-anxiety observers were less inclined to approach happy crowds.
Crowd trials
Scrambled face
p < 0.001
Angry vs. HappySingle
LVF vs. RVF (visual field of presentation of angry stimuli)Crowd Single
Amygdala
pSTS
PHC
TPJ
Precuneus
RL RL
RL RL [1] Haberman, J., & Whitney, D. (2007). Rapid extraction of mean emotion and gender from sets of faces. Current Biology, 17, R751-3. [2] Ekman, P., & Friesen, W. V. (1976). Pictures of facial affect. Consulting Psychologists Press; Palo Alto, CA. [3] Spielberger, C. D., Gorsuch, R. L., Lushene, R., Vagg, P. R., & Jacobs, G. A. (1983). Manual for the State-Trait Anxiety Inventory. Palo Alto, CA: Consulting Psychologists Press.
Temporal pole
Amygdala