Andreas Roepstorff

Functional and Structural Correlates of Social Influence in the Human Brain

Research output: Contribution to conferencePosterResearchpeer-review

  • Daniel Campbell-Meiklejohn, Denmark
  • Dominik Bach, University College London, United Kingdom
  • Ryota Kanai, University College London, United Kingdom
  • Bahador Bahrami, University College London, United Kingdom
  • Andreas Roepstorff
  • Ray Dolan, University College London, United Kingdom
  • Chris D Frith
Values that one associates with available options, from foods to political candidates, help to guide our choices and behaviour. These values can be updated by the preferences expressed by other people as much as by independent experience with the options. However, some people conform their values more than others. Across different contexts, the tendency to conform is a fairly stable trait. This stability suggests that it may have stable anatomical and physiological correlates. We tested for relationships between the tendency to conform values and: a) functional responses to social conflict; b) functional responses reflecting social influence on object value; and c) grey matter volume (GM) the orbitofrontal cortex (OFC) – an area not available to functional analysis in our study but clearly involved in social conduct and value learning.
Prior to testing, 28 healthy subjects provided names of twenty pieces of music that they would like to own, but did not own yet. On test day, subjects rated each submitted song for desirability, from 1 (low) to 10 (high). Next, subjects were told that two music 'reviewers', of whom subjects had read descriptions and rated as capable of choosing enjoyable music, had listened to each song. Subjects performed the task in Figure 1 in the scanner. During a trial, subjects indicated their preference, given a choice of a song they had submitted and an alternative song they had never heard. Subjects were then told which song reviewers preferred before randomly receiving a token for one of the songs. Each submitted song was evaluated against 6 alternative songs. After the task, subjects rerated their desire for each song. The linear relationship between change in desire and reviewer preference for songs (times preferred - times not preferred) was measured. The resulting standardized coefficient, Binf (M .091, SD .17), provided a measure of social influence on song value (conformity) for each subject.

Binf was used as a between-subject regressor in the group fMRI analysis of the contrasts of (a) disagreement (vs. agreement) with reviewers; and (b) social influence on neural value signals (the interaction between object reward (receiving preferred song vs. receiving other) and reviewer preference (prefer same song vs. prefer other)). In the structural analysis of OFC, Binf was entered as a between-subject regressor (along with age, gender and whole brain GM) in the volumetric based morphology (VBM) contrast of GM in OFC. OFCGM found to correlate (FWE corrected, P >.05) with Binf, was used as a regressor in the group fMRI contrasts of (a) and (b) to investigate the relationship between structure and activity during conformity.
Original languageEnglish
Publication year13 Jun 2012
Publication statusPublished - 13 Jun 2012

See relations at Aarhus University Citationformats

ID: 45896699