Neuroforecasting online engagement with nature images
Poster No:
718
Submission Type:
Abstract Submission
Authors:
Tara Srirangarajan1, Nik Sawe1, Tierney Thys2, Cynthia Wu1, Brian Knutson1
Institutions:
1Stanford University, Stanford, CA, 2California Academy of Sciences, San Francisco, CA
First Author:
Co-Author(s):
Introduction:
Visual images can facilitate online engagement and support for endangered species, but the mechanisms supporting their impact remains unclear. By combining behavior, neuroimaging, and surveys, we sought to examine how images depicting endangered species encourage engagement on a popular social media platform (i.e., Instagram.com).
Specifically, we leveraged a neuroforecasting approach by using brain activity prior to choice not only to predict subsequent choice in individuals, but also to forecast aggregate choice in other groups (1). Prior research has shown that group neural activity can forecast aggregate demand in different markets ranging from music sales (2) to the effectiveness of health advertising campaigns (3). A consistent set of brain regions whose activity has been associated with affect and motivation have been implicated in previous neuroforecasting literature. According to an Affect-Integration-Motivation (or AIM) framework, Nucleus Accumbens (or NAcc) activity is associated with gain anticipation and predicts approach behavior, Anterior Insula (or AIns) activity is associated with loss anticipation and predicts avoidance behavior, and Medial PreFrontal Cortex (or MPFC) activity is associated with value integration and identity and predicts approach behavior (4). We predicted that activity in these regions (i.e., +NAcc, –AIns, +MPFC) would predict individual engagement with nature images, and that group activity in a subset of these regions might also forecast aggregate engagement.
Specifically, we leveraged a neuroforecasting approach by using brain activity prior to choice not only to predict subsequent choice in individuals, but also to forecast aggregate choice in other groups (1). Prior research has shown that group neural activity can forecast aggregate demand in different markets ranging from music sales (2) to the effectiveness of health advertising campaigns (3). A consistent set of brain regions whose activity has been associated with affect and motivation have been implicated in previous neuroforecasting literature. According to an Affect-Integration-Motivation (or AIM) framework, Nucleus Accumbens (or NAcc) activity is associated with gain anticipation and predicts approach behavior, Anterior Insula (or AIns) activity is associated with loss anticipation and predicts avoidance behavior, and Medial PreFrontal Cortex (or MPFC) activity is associated with value integration and identity and predicts approach behavior (4). We predicted that activity in these regions (i.e., +NAcc, –AIns, +MPFC) would predict individual engagement with nature images, and that group activity in a subset of these regions might also forecast aggregate engagement.
Methods:
37 healthy right-handed English-speaking adults with pre-existing Instagram accounts participated in the neuroimaging study. FMRI data were acquired using a 3.0 T General Electric MRI scanner equipped with a 32-channel head coil at the Stanford Center for Cognitive and Neurobiological Imaging. During FMRI scanning, participants viewed 56 photographs selected from the Instagram feed of a prominent environmentally-focused publication (i.e., National Geographic Magazine). Photos sampled wildlife images selected from a pool of 888 total images posted over the span of three months. 14 images were preselected from each quartile of engagement (e.g., the most and least popular quarter of photos). Trials first included presentation of an image (4 sec), next a question about liking (2 sec), followed by a choice prompt (Y/N counterbalanced left/right; 4 sec), then a question about donation (2 sec) followed by a choice prompt (Y/N counterbalanced left/right; 4 sec), and finally a variable intertrial interval indicated by a central fixation cross (2-6 sec). To maintain incentive compatibility, one liking decision and one donation decision were randomly selected to count as binding at the conclusion of the experiment.
For anatomically-targeted data analyses, neural Volumes Of Interest (VOIs) were specified in regions highlighted in the AIM framework (4). Specifically, spherical Volumes Of Interest (VOIs; 8 mm in diameter) were centered on bilateral foci in the Nucleus Accumbens (NAcc; TC: ±10, 12, –2), Anterior Insula (AIns; TC: ±34, 24, –4), and Medial PreFrontal Cortex (MPFC; TC: ±4, 45, 0). Activity within each VOI was averaged across voxels and then averaged bilaterally to derive activity time courses.
For anatomically-targeted data analyses, neural Volumes Of Interest (VOIs) were specified in regions highlighted in the AIM framework (4). Specifically, spherical Volumes Of Interest (VOIs; 8 mm in diameter) were centered on bilateral foci in the Nucleus Accumbens (NAcc; TC: ±10, 12, –2), Anterior Insula (AIns; TC: ±34, 24, –4), and Medial PreFrontal Cortex (MPFC; TC: ±4, 45, 0). Activity within each VOI was averaged across voxels and then averaged bilaterally to derive activity time courses.
Results:
As predicted, findings indicated that activity in brain circuits associated with anticipatory affect (i.e., the NAcc and MPFC) predicted individual liking and donations (NAcc: Z=2.11, p < 0.001; MPFC: Z=5.21, p < 0.001). Group brain activity in an integrative part of this circuit (MPFC) also forecast engagement on Instagram (t=2.55, p < 0.001) (Figure 1).
Conclusions:
To examine how images of endangered species elicit engagement, we combined behavioral probes, a neuroimaging experiment, and representative surveys. These findings not only extend neuroforecasting to online engagement with images of endangered species, but also offer the possibility that neural data might reveal which image features drive online engagement.
Emotion, Motivation and Social Neuroscience:
Emotion and Motivation Other 1
Higher Cognitive Functions:
Decision Making 2
Higher Cognitive Functions Other
Modeling and Analysis Methods:
Activation (eg. BOLD task-fMRI)
Novel Imaging Acquisition Methods:
BOLD fMRI
Keywords:
Emotions
FUNCTIONAL MRI
Sub-Cortical
1|2Indicates the priority used for review
Provide references using author date format
2. Berns, G. S. (2012), 'A neural predictor of cultural popularity,' Journal of Consumer Psychology 22, 154–160.
3. Falk, E. B. (2012), 'From Neural Responses to Population Behavior: Neural Focus Group Predicts Population-Level Media Effects,' Psychological Science 23, 439–45.
4. Samanez-Larkin, G. R. (2015), 'Decision making in the ageing brain: changes in affective and motivational circuits,' Nature Reviews Neuroscience, 16(5), 278–289.