Format

Send to

Choose Destination
Brain Imaging Behav. 2019 Jan 31. doi: 10.1007/s11682-019-00035-5. [Epub ahead of print]

Meta-analysis of the moral brain: patterns of neural engagement assessed using multilevel kernel density analysis.

Author information

1
Psychology Department, University of New Mexico, Albuquerque, NM, USA. fedesj@nih.gov.
2
Mind Research Network and Lovelace Biomedical and Environmental Institute (LBERI), 1101 Yale Blvd NE, Albuquerque, NM, 87106, USA. fedesj@nih.gov.
3
Psychology Department, University of New Mexico, Albuquerque, NM, USA.
4
Mind Research Network and Lovelace Biomedical and Environmental Institute (LBERI), 1101 Yale Blvd NE, Albuquerque, NM, 87106, USA.

Abstract

The neuroimaging literature in moral cognition has rapidly developed in the last decade with more than 200 publications on the topic. Neuroimaging based models generally agree that limbic regions work with medial prefrontal and temporal regions during moral processing to integrate emotional, social, and cognitive elements into decision-making. However, no quantitative work has been done examining neural response across types of moral cognition tasks. This paper uses Multilevel Kernel Density Analysis (MKDA) to conduct neuroimaging meta-analyses of the moral cognitive literature. MKDA replicated previous findings of the neural correlates of moral cognition: the left amygdala, medial prefrontal cortex, bilateral temporoparietal junction, and posterior cingulate. Random forest algorithms classified neural features as belonging to simple/utilitarian moral dilemmas, explicit/implicit moral tasks, and word/picture moral stimuli tasks; in combination with univariate contrast analyses, these results indicated a distinct pattern of processing for each of the members of these paradigm pairs. Overall, the results emphasize that the task selected for use in a moral cognition neuroimaging study is vital for the elicitation and interpretation of results. It also replicates and re-establishes the neural basis for moral processing, especially important in light of implementation errors in previous meta-analysis.

KEYWORDS:

MKDA; Machine learning; Meta-analysis; Moral; fMRI

PMID:
30706370
DOI:
10.1007/s11682-019-00035-5

Supplemental Content

Full text links

Icon for Springer
Loading ...
Support Center