Elsevier

NeuroImage

Volume 112, 15 May 2015, Pages 138-151
NeuroImage

Functional brain networks underlying detection and integration of disconfirmatory evidence

https://doi.org/10.1016/j.neuroimage.2015.02.043Get rights and content

Highlights

  • Integration of disconfirmatory evidence requires alerting and integration phases.

  • We identified sequentially-active brain networks associated with these processes.

  • The salience network activated at the onset of the disconfirmatory evidence.

  • A second frontoparietal brain network activated during integration of the evidence.

  • Integration of disconfirmatory evidence recruits sequentially-active networks.

Abstract

Processing evidence that disconfirms a prior interpretation is a fundamental aspect of belief revision, and has clear social and clinical relevance. This complex cognitive process requires (at minimum) an alerting stage and an integration stage, and in the current functional magnetic resonance imaging (fMRI) study, we used multivariate analysis methodology on two datasets in an attempt to separate these sequentially-activated cognitive stages and link them to distinct functional brain networks. Thirty-nine healthy participants completed one of two versions of an evidence integration experiment involving rating two consecutive animal images, both of which consisted of two intact images of animal faces morphed together at different ratios (e.g., 70/30 bird/dolphin followed by 10/90 bird/dolphin). The two versions of the experiment differed primarily in terms of stimulus presentation and timing, which facilitated functional interpretation of brain networks based on differences in the hemodynamic response shapes between versions. The data were analyzed using constrained principal component analysis for fMRI (fMRI-CPCA), which allows distinct, simultaneously active task-based networks to be separated, and these were interpreted using both temporal (task-based hemodynamic response shapes) and spatial (dominant brain regions) information. Three networks showed increased activity during integration of disconfirmatory relative to confirmatory evidence: (1) a network involved in alerting to the requirement to revise an interpretation, identified as the salience network (dorsal anterior cingulate cortex and bilateral insula); (2) a sensorimotor response-related network (pre- and post-central gyri, supplementary motor area, and thalamus); and (3) an integration network involving rostral prefrontal, orbitofrontal and posterior parietal cortex. These three networks were staggered in their peak activity (alerting, responding, then integrating), but at certain time points (e.g., 17 s after trial onset) the hemodynamic responses associated with all three networks were simultaneously active. These findings highlight distinct cognitive processes and corresponding functional brain networks underlying stages of disconfirmatory evidence integration, and demonstrate the power of multivariate and multi-experiment methodology in cognitive neuroscience.

Section snippets

Participants

Participants were 39 healthy volunteers (Version 1: 10 male, 10 female, mean age = 24.90, SD = 6.87; Version 2: 9 male, 10 female, mean age = 26.84, SD = 7.34), most of which were native English speakers (Version 1: 17 participants; Version 2: 15 participants). Non-native English speakers had been using English daily for at least the past five years and responded accurately to questions about the consent form designed to confirm their ability to read and understand English. All participants were

Results

Inspection of the scree plot of singular values (Cattell, 1966, Cattell and Vogelmann, 1977) suggested that five components should be extracted. The percentages of task-related variance accounted for by each rotated component were 10.88%, 10.43%, 9.00%, 7.64%, and 6.20%, for Components 1 to 5, respectively. For Component 3,1

Discussion

In the current study we used multivariate methodology on two datasets in an attempt to link two sequential cognitive stages involved in integrating disconfirmatory evidence to distinct functional brain networks. Three functional networks showed greater intensity (i.e., increased activations and/or increased deactivations) during integration of disconfirmatory relative to confirmatory evidence for both experiment versions. In order of peak timing (see Figs. 5B, 4B, and 2B, respectively), these

Acknowledgments

This project was supported by a Scholar Award from the Michael Smith Foundation for Health Research (MSFHR; CI-SCH-00073), a New Investigator Award from the Canadian Institutes of Health Research (CIHR; MMS8770) to TSW, as well as CIHR Doctoral Research Awards for KML (DSZ-128637) and PDM (DPO-128616). Operating costs were supported by a grant to TSW from the British Columbia Schizophrenia Society (BCSS; formerly Mind Foundation of BC). The authors acknowledge the UBC High Field Magnetic

References (51)

  • M. Annett

    A classification of hand preference by association analysis

    Br. J. Psychol.

    (1970)
  • T.E.J. Behrens et al.

    Learning the value of information in an uncertain world

    Nat. Neurosci.

    (2007)
  • J.R. Binder et al.

    Where is the semantic system? A critical review and meta-analysis of 120 functional neuroimaging studies

    Cereb. Cortex

    (2009)
  • R.L. Buckner et al.

    The brain's default network

    Ann. N. Y. Acad. Sci.

    (2008)
  • R.L. Buckner et al.

    The organization of the human cerebellum estimated by intrinsic functional connectivity

    J. Neurophysiol.

    (2011)
  • P.W. Burgess et al.

    The gateway hypothesis of rostral prefrontal cortex (area 10) function

  • S.C. Cameron et al.

    Anterior cingulate cortex and conflict detection: an update of theory and data

    Cogn. Affect. Behav. Neurosci.

    (2007)
  • R.B. Cattell

    The scree test for the number of factors

    Multivar. Behav. Res.

    (1966)
  • R.B. Cattell et al.

    A comprehensive trial of the scree and kg criteria for determining the number of factors

    Multivar. Behav. Res.

    (1977)
  • E.Y. Choi et al.

    The organization of the human striatum estimated by intrinsic functional connectivity

    J. Neurophysiol.

    (2012)
  • K. Christoff et al.

    Evaluating self-generated information: anterior prefrontal contributions to human cognition

    Behav. Neurosci.

    (2003)
  • M. Coltheart

    The neuropsychology of delusions

    Ann. N. Y. Acad. Sci.

    (2010)
  • M. d'Acremont et al.

    The human brain encodes event frequencies while forming subjective beliefs

    J. Neurosci.

    (2013)
  • G.I. de Zubicaray et al.

    Differential processing of thematic and categorical conceptual relations in spoken word production

    J. Exp. Psychol. Gen.

    (2013)
  • T. Egner

    Surprise! A unifying model of dorsal anterior cingulate function?

    Nat. Neurosci.

    (2011)
  • Cited by (29)

    • Altered activity in functional brain networks involved in lexical decision making in bipolar disorder: An fMRI case-control study

      2022, Journal of Affective Disorders
      Citation Excerpt :

      Supplementary Table S3 shows the correspondence of activity peaks in Component 3 with anatomical descriptions, BAs and resting state networks. Supplementary Fig. S3 shows slice comparisons between the current study and previous studies (Lavigne et al., 2015, 2020) in our lab for Component 3. Activation in this network included left-dominant pre- and post-central gyri (BAs 3, 6), and juxtapositional lobule cortex (BA 6), which are often motor response network regions when performing a right-handed response.

    • Functional brain networks underlying evidence integration and delusional ideation

      2020, Schizophrenia Research
      Citation Excerpt :

      This network's deactivating regions correspond to the well-established DMN (Buckner et al., 2008). Activating regions (especially subthreshold) were similar to the VsAN (see component 3), as was observed in our two previous studies (Lavigne et al., 2019; Lavigne et al., 2015). There was a significant Post-stimulus Time × Condition interaction, F(9,360) = 3.66, p < .001, due to increased disconfirm versus confirm activity at 9 s, 11 s, and 13 s post-stimulus (all ps < .05), which corresponds to the peak of the HDR (Fig. 2B).

    • Functional brain networks involved in lexical decision

      2020, Brain and Cognition
      Citation Excerpt :

      This will capture all HDR fluctuations that are consistent over trials, including the canonical HDR and any delayed HDRs. Task-based network approaches to the analysis of fMRI data have demonstrated that attentional/response processes typically fall onto separate brain networks (Goghari, Sanford, Spilka, & Woodward, 2017; Lavigne, Metzak, & Woodward, 2015a; Woodward, Feredoes, Metzak, Takane, & Manoach, 2013), and the HDR shape associated with each (specifically, the HDR to experimental manipulations) can be inspected alongside the anatomical distribution to determine the function of each network. That is to say, comparison between task-based functional brain networks with respect to the HDR shapes, and the experimental conditions to which the networks respond, is instructive for understanding the function of each brain networks involved in LD.

    View all citing articles on Scopus
    View full text