Classification of Risky Decision-Making from Adolescents through Real-Time fMRI Neurofeedback
Poster No:
709
Submission Type:
Abstract Submission
Authors:
Dong-Youl Kim1, Mark Orloff2, Jonathan Lisinski1, Jungmeen Kim-Spoon3, Stephen LaConte1,4, Pearl Chiu1,3, Brooks Casas1,3
Institutions:
1Fralin Biomedical Research Institute at VTC, Virginia Tech, Roanoke, VA, 2The Center for Mind and Brain, University of California, Davis, Davis, CA, 3Department of Psychology, Virginia Tech, Blacksburg, VA, 4Department of Biomedical Engineering and Mechanics, Virginia Tech, Blacksburg, VA
First Author:
Co-Author(s):
Stephen LaConte, PhD
Fralin Biomedical Research Institute at VTC, Virginia Tech|Department of Biomedical Engineering and Mechanics, Virginia Tech
Roanoke, VA|Blacksburg, VA
Fralin Biomedical Research Institute at VTC, Virginia Tech|Department of Biomedical Engineering and Mechanics, Virginia Tech
Roanoke, VA|Blacksburg, VA
Pearl Chiu
Fralin Biomedical Research Institute at VTC, Virginia Tech|Department of Psychology, Virginia Tech
Roanoke, VA|Blacksburg, VA
Fralin Biomedical Research Institute at VTC, Virginia Tech|Department of Psychology, Virginia Tech
Roanoke, VA|Blacksburg, VA
Brooks Casas
Fralin Biomedical Research Institute at VTC, Virginia Tech|Department of Psychology, Virginia Tech
Roanoke, VA|Blacksburg, VA
Fralin Biomedical Research Institute at VTC, Virginia Tech|Department of Psychology, Virginia Tech
Roanoke, VA|Blacksburg, VA
Introduction:
Previous fMRI studies report changes of brain activity, connectivity, as well as intrinsic networks associated with risky decision-making (Chung et al., 2015; Lauharatanahirun et al., 2018). The application of whole-brain multivariate analysis may thus modulate brain functions in a more effective manner than within-region neurofeedback methods for targeting decision-making processes (Kim et al., 2024; LaConte et al., 2005; LaConte et al., 2007). This study aims to examine real-time fMRI (rtfMRI) neurofeedback based on whole-brain activation classification to regulate adolescents' risky decision-making.
Methods:
To develop a whole-brain classifier for use in a rtfMRI setting, a dataset including 143 participants (non-rtfMRI) was analyzed to discriminate between riskier and safer choices. Condition-wise beta images were estimated using a general linear model. A nu-support vector machine with a non-linear kernel was adopted to train the classifier in a 10-fold nested cross-validation scheme using individual neural patterns. For online analysis, the percentage BOLD signal was estimated and the distance to the model hyperplane was presented via a slider bar. As shown in Figure 1, the experiment consisted of 5 sessions, with the first and last sessions consisting of 72 decision trials without feedback. In the second through fourth sessions, six runs were completed including two non-real-time runs (36 trials each occurring at the beginning and end of the session) and four real-time runs (18 trials each). For offline analysis using 9 subjects, the percentage BOLD signal was calculated to evaluate model performance. In addition, the condition-wise (i.e., risky or safer choices) beta map for each subject was estimated and then spatial correlation coefficients between beta maps and reference images (i.e., weight patterns from model) were measured.
Results:
The model used to provide neurofeedback had a test accuracy of 79.93% (Figure 2a). The weight spatial patterns from a whole-brain classifier from 143 participants were identified across brain regions including bilateral insula, medial orbitofrontal cortex, medial superior frontal gyrus, caudate, superior temporal pole, dorsal lateral prefrontal cortex, anterior/posterior cingulate cortex, and inferior parietal lobule (Figure 2b). For offline analyses using five sessions including rtfMRI neurofeedback runs, high accuracy (>70%) and increased performance across the five sessions was observed via linear regression (p < 10-4; Figure 2c). Significant increased correlation was observed using risky choice related patterns between session 1 and 5 (corrected p < 0.05) but not using safe choice related patterns between session 1 and 5 (Figure 2d).
Conclusions:
These results suggest the potential for successfully porting group data into a real-time fMRI setting to modulate brain processes associated with risky decision-making. The effect through repeated training on decision-making could lead to changes in brain patterns enhancing distinctness of risky and safe decision-making processes.
Emotion, Motivation and Social Neuroscience:
Social Neuroscience Other
Higher Cognitive Functions:
Decision Making 1
Modeling and Analysis Methods:
Activation (eg. BOLD task-fMRI)
Classification and Predictive Modeling
Multivariate Approaches 2
Keywords:
Other - Real-time fMRI; neurofeedback; decision-making process; support vector machine; classification.
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Are you Internal Review Board (IRB) certified? Please note: Failure to have IRB, if applicable will lead to automatic rejection of abstract.
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Which processing packages did you use for your study?
Provide references using APA citation style.
[2] Lauharatanahirun et al., Neural Correlates of Risk Processing Among Adolescents: Influences of Parental Monitoring and Household Chaos. Child Dev. (2018), 89(3), 784–796.
[3] Kim et al., Regulation of craving for real-time fMRI neurofeedback based on individual classification. Philos. Trans. R. Soc. B (2024), 379(1915), 20230094.
[4] LaConte et al., Support vector machines for temporal classification of block design fMRI data. NeuroImage (2005), 26(2), 317-329.
[5] LaConte et al., Real-time fMRI using brain-state classification. Hum. Brain Map. (2007), 28(10), 1033-1044.
This work was supported in part by the National Institutes of Health (DA061024 to BC, JK; MH122948 to BC; DA051573 to BC, PC).