Accounting for dynamic fluctuations across time when examining fMRI test-retest reliability

Analysis of a reward paradigm in the EMBARC study

Henry W. Chase, Jay C. Fournier, Tsafrir Greenberg, Jorge R. Almeida, Richelle Stiffler, Carlos R. Zevallos, Haris Aslam, Crystal Cooper, Thilo Deckersbach, Sarah Weyandt, Phillip Adams, Marisa Toups, Tom Carmody, Maria A. Oquendo, Scott Peltier, Maurizio Fava, Patrick J. McGrath, Myrna Weissman, Ramin Parsey, Melvin G. McInnis & 3 others Benji Kurian, Madhukar H. Trivedi, Mary L. Phillips

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

Longitudinal investigation of the neural correlates of reward processing in depression may represent an important step in defining effective biomarkers for antidepressant treatment outcome prediction, but the reliability of reward-related activation is not well understood. Thirty-seven healthy control participants were scanned using fMRI while performing a reward-related guessing task on two occasions, approximately one week apart. Two main contrasts were examined: right ventral striatum (VS) activation fMRI BOLD signal related to signed prediction errors (PE) and reward expectancy (RE). We also examined bilateral visual cortex activation coupled to outcome anticipation. Significant VS PE-related activity was observed at the first testing session, but at the second testing session, VS PE-related activation was significantly reduced. Conversely, significant VS RE-related activity was observed at time 2 but not time 1. Increases in VS RE-related activity from time 1 to time 2 were significantly associated with decreases in VS PE-related activity from time 1 to time 2 across participants. Intraclass correlations (ICCs) in VS were very low. By contrast, visual cortex activation had much larger ICCs, particularly in individuals with high quality data. Dynamic changes in brain activation are widely predicted, and failure to account for these changes could lead to inaccurate evaluations of the reliability of functional MRI signals. Conventional measures of reliability cannot distinguish between changes specified by algorithmic models of neural function and noisy signal. Here, we provide evidence for the former possibility: reward-related VS activations follow the pattern predicted by temporal difference models of reward learning but have low ICCs.

Original languageEnglish (US)
Article numbere0126326
JournalPLoS One
Volume10
Issue number5
DOIs
StatePublished - May 11 2015

Fingerprint

Reliability analysis
Reward
Reproducibility of Results
Chemical activation
Magnetic Resonance Imaging
prediction
testing
Visual Cortex
antidepressants
biomarkers
learning
Testing
Biomarkers
Ventral Striatum
brain
Antidepressive Agents
Brain
Healthy Volunteers
Learning
Depression

ASJC Scopus subject areas

  • Agricultural and Biological Sciences(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Medicine(all)

Cite this

Chase, H. W., Fournier, J. C., Greenberg, T., Almeida, J. R., Stiffler, R., Zevallos, C. R., ... Phillips, M. L. (2015). Accounting for dynamic fluctuations across time when examining fMRI test-retest reliability: Analysis of a reward paradigm in the EMBARC study. PLoS One, 10(5), [e0126326]. https://doi.org/10.1371/journal.pone.0126326

Accounting for dynamic fluctuations across time when examining fMRI test-retest reliability : Analysis of a reward paradigm in the EMBARC study. / Chase, Henry W.; Fournier, Jay C.; Greenberg, Tsafrir; Almeida, Jorge R.; Stiffler, Richelle; Zevallos, Carlos R.; Aslam, Haris; Cooper, Crystal; Deckersbach, Thilo; Weyandt, Sarah; Adams, Phillip; Toups, Marisa; Carmody, Tom; Oquendo, Maria A.; Peltier, Scott; Fava, Maurizio; McGrath, Patrick J.; Weissman, Myrna; Parsey, Ramin; McInnis, Melvin G.; Kurian, Benji; Trivedi, Madhukar H.; Phillips, Mary L.

In: PLoS One, Vol. 10, No. 5, e0126326, 11.05.2015.

Research output: Contribution to journalArticle

Chase, HW, Fournier, JC, Greenberg, T, Almeida, JR, Stiffler, R, Zevallos, CR, Aslam, H, Cooper, C, Deckersbach, T, Weyandt, S, Adams, P, Toups, M, Carmody, T, Oquendo, MA, Peltier, S, Fava, M, McGrath, PJ, Weissman, M, Parsey, R, McInnis, MG, Kurian, B, Trivedi, MH & Phillips, ML 2015, 'Accounting for dynamic fluctuations across time when examining fMRI test-retest reliability: Analysis of a reward paradigm in the EMBARC study', PLoS One, vol. 10, no. 5, e0126326. https://doi.org/10.1371/journal.pone.0126326
Chase, Henry W. ; Fournier, Jay C. ; Greenberg, Tsafrir ; Almeida, Jorge R. ; Stiffler, Richelle ; Zevallos, Carlos R. ; Aslam, Haris ; Cooper, Crystal ; Deckersbach, Thilo ; Weyandt, Sarah ; Adams, Phillip ; Toups, Marisa ; Carmody, Tom ; Oquendo, Maria A. ; Peltier, Scott ; Fava, Maurizio ; McGrath, Patrick J. ; Weissman, Myrna ; Parsey, Ramin ; McInnis, Melvin G. ; Kurian, Benji ; Trivedi, Madhukar H. ; Phillips, Mary L. / Accounting for dynamic fluctuations across time when examining fMRI test-retest reliability : Analysis of a reward paradigm in the EMBARC study. In: PLoS One. 2015 ; Vol. 10, No. 5.
@article{59adaf31746f4ce3bc6d92fe60094cfe,
title = "Accounting for dynamic fluctuations across time when examining fMRI test-retest reliability: Analysis of a reward paradigm in the EMBARC study",
abstract = "Longitudinal investigation of the neural correlates of reward processing in depression may represent an important step in defining effective biomarkers for antidepressant treatment outcome prediction, but the reliability of reward-related activation is not well understood. Thirty-seven healthy control participants were scanned using fMRI while performing a reward-related guessing task on two occasions, approximately one week apart. Two main contrasts were examined: right ventral striatum (VS) activation fMRI BOLD signal related to signed prediction errors (PE) and reward expectancy (RE). We also examined bilateral visual cortex activation coupled to outcome anticipation. Significant VS PE-related activity was observed at the first testing session, but at the second testing session, VS PE-related activation was significantly reduced. Conversely, significant VS RE-related activity was observed at time 2 but not time 1. Increases in VS RE-related activity from time 1 to time 2 were significantly associated with decreases in VS PE-related activity from time 1 to time 2 across participants. Intraclass correlations (ICCs) in VS were very low. By contrast, visual cortex activation had much larger ICCs, particularly in individuals with high quality data. Dynamic changes in brain activation are widely predicted, and failure to account for these changes could lead to inaccurate evaluations of the reliability of functional MRI signals. Conventional measures of reliability cannot distinguish between changes specified by algorithmic models of neural function and noisy signal. Here, we provide evidence for the former possibility: reward-related VS activations follow the pattern predicted by temporal difference models of reward learning but have low ICCs.",
author = "Chase, {Henry W.} and Fournier, {Jay C.} and Tsafrir Greenberg and Almeida, {Jorge R.} and Richelle Stiffler and Zevallos, {Carlos R.} and Haris Aslam and Crystal Cooper and Thilo Deckersbach and Sarah Weyandt and Phillip Adams and Marisa Toups and Tom Carmody and Oquendo, {Maria A.} and Scott Peltier and Maurizio Fava and McGrath, {Patrick J.} and Myrna Weissman and Ramin Parsey and McInnis, {Melvin G.} and Benji Kurian and Trivedi, {Madhukar H.} and Phillips, {Mary L.}",
year = "2015",
month = "5",
day = "11",
doi = "10.1371/journal.pone.0126326",
language = "English (US)",
volume = "10",
journal = "PLoS One",
issn = "1932-6203",
publisher = "Public Library of Science",
number = "5",

}

TY - JOUR

T1 - Accounting for dynamic fluctuations across time when examining fMRI test-retest reliability

T2 - Analysis of a reward paradigm in the EMBARC study

AU - Chase, Henry W.

AU - Fournier, Jay C.

AU - Greenberg, Tsafrir

AU - Almeida, Jorge R.

AU - Stiffler, Richelle

AU - Zevallos, Carlos R.

AU - Aslam, Haris

AU - Cooper, Crystal

AU - Deckersbach, Thilo

AU - Weyandt, Sarah

AU - Adams, Phillip

AU - Toups, Marisa

AU - Carmody, Tom

AU - Oquendo, Maria A.

AU - Peltier, Scott

AU - Fava, Maurizio

AU - McGrath, Patrick J.

AU - Weissman, Myrna

AU - Parsey, Ramin

AU - McInnis, Melvin G.

AU - Kurian, Benji

AU - Trivedi, Madhukar H.

AU - Phillips, Mary L.

PY - 2015/5/11

Y1 - 2015/5/11

N2 - Longitudinal investigation of the neural correlates of reward processing in depression may represent an important step in defining effective biomarkers for antidepressant treatment outcome prediction, but the reliability of reward-related activation is not well understood. Thirty-seven healthy control participants were scanned using fMRI while performing a reward-related guessing task on two occasions, approximately one week apart. Two main contrasts were examined: right ventral striatum (VS) activation fMRI BOLD signal related to signed prediction errors (PE) and reward expectancy (RE). We also examined bilateral visual cortex activation coupled to outcome anticipation. Significant VS PE-related activity was observed at the first testing session, but at the second testing session, VS PE-related activation was significantly reduced. Conversely, significant VS RE-related activity was observed at time 2 but not time 1. Increases in VS RE-related activity from time 1 to time 2 were significantly associated with decreases in VS PE-related activity from time 1 to time 2 across participants. Intraclass correlations (ICCs) in VS were very low. By contrast, visual cortex activation had much larger ICCs, particularly in individuals with high quality data. Dynamic changes in brain activation are widely predicted, and failure to account for these changes could lead to inaccurate evaluations of the reliability of functional MRI signals. Conventional measures of reliability cannot distinguish between changes specified by algorithmic models of neural function and noisy signal. Here, we provide evidence for the former possibility: reward-related VS activations follow the pattern predicted by temporal difference models of reward learning but have low ICCs.

AB - Longitudinal investigation of the neural correlates of reward processing in depression may represent an important step in defining effective biomarkers for antidepressant treatment outcome prediction, but the reliability of reward-related activation is not well understood. Thirty-seven healthy control participants were scanned using fMRI while performing a reward-related guessing task on two occasions, approximately one week apart. Two main contrasts were examined: right ventral striatum (VS) activation fMRI BOLD signal related to signed prediction errors (PE) and reward expectancy (RE). We also examined bilateral visual cortex activation coupled to outcome anticipation. Significant VS PE-related activity was observed at the first testing session, but at the second testing session, VS PE-related activation was significantly reduced. Conversely, significant VS RE-related activity was observed at time 2 but not time 1. Increases in VS RE-related activity from time 1 to time 2 were significantly associated with decreases in VS PE-related activity from time 1 to time 2 across participants. Intraclass correlations (ICCs) in VS were very low. By contrast, visual cortex activation had much larger ICCs, particularly in individuals with high quality data. Dynamic changes in brain activation are widely predicted, and failure to account for these changes could lead to inaccurate evaluations of the reliability of functional MRI signals. Conventional measures of reliability cannot distinguish between changes specified by algorithmic models of neural function and noisy signal. Here, we provide evidence for the former possibility: reward-related VS activations follow the pattern predicted by temporal difference models of reward learning but have low ICCs.

UR - http://www.scopus.com/inward/record.url?scp=84930624776&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84930624776&partnerID=8YFLogxK

U2 - 10.1371/journal.pone.0126326

DO - 10.1371/journal.pone.0126326

M3 - Article

VL - 10

JO - PLoS One

JF - PLoS One

SN - 1932-6203

IS - 5

M1 - e0126326

ER -