Introduction to Functional MRI

This note overviews the fundamentals of fMRI, from physiology to analysis. It is not intended as an exhaustive list of all the classics and does not contain some of the seminal early studies. It provides recent overviews of the key issues and interested readers should consult the original references provided within these articles.

General references

• A lay video introduction
• Video lectures on fMRI fundamentals by Rebecca Saxe
• Some resources on what it's like to be scanned
• The Huettel et al. book is excellent
• Another book focused on analysis is by Poldrack et al.
• A more technical treatment focused on SPM is provided by Friston et al. Here is a link to the free E-version of Human Brain Function
• A Coursera course by Wager and Lindquist (Part 1)
• A Coursera course by Wager and Lindquist (Part 2)
• A lot of great resources are available on the MRC CBU wiki
• The FSL course is excellent and slides can be found here
• An NIH syllabus on fMRI (Main page)
• An NIH syllabus on fMRI (Summer 2019)
• List of resources for learning about fMRI
• An SPM course
• Introductory lecture for fMRI
• A nice reading list as an intro to fMRI
• 54 lectures on different aspects of imaging and cognitive neuroscience
• Nice intro to some basic aspects of MRI
• A great blog on various aspects of fMRI processing and analysis
• Nice GitHub repo to help people understand the basis of fMRI signal through code
• An edited book by M. Fillipi
• Good overview of physiology, physics, and analysis of fMRI
• Guideline on how to read an fMRI paper
• Primer on the GLM for MRI

History of fMRI

This is a series of articles contributed as part of a NeuroImage special issue on 20 years of fMRI that provide great insights into historical developments and the basis of the signal:

Author	Journal	Year	Title
Bandettini et al.	NeuroImage	2012	Overview of 20 years of fMRI research
Bandettini et al.	NeuroImage	2012	First fMRI experiment
Kwong et al.	NeuroImage	2012	The first fMRI experiment
Ogawa et al.	NeuroImage	2012	Discovery of BOLD effect
Turner et al.	NeuroImage	2012	History of fMRI developments
Ugurbil et al.	NeuroImage	2012	First fMRI experiment

Physiological basis of the BOLD signal

Some key articles from Logothetis, who pioneered studies of the neuronal basis of BOLD:

• A list of physiological processes contributing to the fMRI signal
• A list of possible physiological causes of group differences in fMRI

Metabolic basis of the BOLD signal:

Author	Journal	Year	Title
Fox	NeuroImage	2012	The coupling controversy
Raichle and Mintun	Annu. Rev. Neurosci.	2006	Brain Work and Brain Imaging
Hyder and Rothman	NeuroImage	2012	Quantitative fMRI and oxidative neuroenergetics
Magistrett and Allaman	Neuron	2015	A Cellular Perspective on Brain Energy Metabolism and Functional Imaging

Modelling The HRF: BOLD measures neural activity convolved with a haemodynamic response function. The HRF is fundamental to all fMRI analysis and can be modelled in different ways. Most software packages issue a canonical shape, but it is useful to be aware of the various issues surrounding the HRF.

Author	Journal	Year	Title
Boynton et al.	NeuroImage	2012	Linear systems analysis of the fMRI signal
Buxton	NeuroImage	2012	Dynamic models of BOLD contrast
Aguirre et al.	NeuroImage	1998	The Variability of Human, BOLD Hemodynamic Responses

• A list of physiological processes contributing to the fMRI signal
• A list of possible physiological causes of group differences in fMRI

Experimental design for task fMRI

When doing task fMRI, experimental design is critical. At a general level, there is a distinction between block-design and event-related designs. There are also various combinations of both and other variants. These articles outline the basics. 

Overviews of different aspects of experimental design in fMRI:

Author	Journal	Year	Title
Clark	NeuroImage	2012	A history of randomized task designs in fMRI
Courtney	NeuroImage	2012	Development of orthogonal task designs in fMRI studies of higher cognition: The NIMH experience
Huettel	NeuroImage	2012	Event-related fMRI in cognition
Liu	NeuroImage	2012	The development of event-related fMRI designs
Petersen and Dubis	NeuroImage	2012	The mixed block/event-related design
Serences	NeuroImage	2004	A comparison of methods for characterizing the event-related BOLD timeseries in rapid fMRI

Design efficiency for event-related studies: an SPM book chapter (Henson, Efficient Experimental Design for fMRI, in SPM by Karl Friston 978-0-12-372560-8) and their website

Optimal task design for fMRI: Durnez et al, bioRxiv 2017, Neurodesign: Optimal experimental designs for task fMRI

Processing and analysis

General Resources

• Excellent set of video tutorials on various aspects of fMRI design and stats
• Some points on slice-timing correction

Basic quality control

• Excellent blog post on understanding image artifacts
• A talk on fMRI QC
• Some advice on checking for basic artifacts and data diagnostics are provided at the MRC CBU page
• A useful overview of checking for data quality is provided in this article (Power, NeuroImage 2017). Although it is focused on resting-state, many of the ideas can apply to task fMRI.

Overviews of the GLM

• Poline and Brett, NeuroImage 2012, The general linear model and fMRI: Does love last forever?

Multiple comparison correction

• Overview of methods: Nichols, NeuroImage 2012, Multiple testing corrections, nonparametric methods, and random field theory
• Setting the primary threshold in cluster-based thresholding: Woo et al, NeuroImage 2014, Cluster-extent based thresholding in fMRI analyses: pitfalls and recommendations
• Problems with cluster-based thresholding:
  • Paper
  • Commentary from OHBM
  • Commentary from Flandin and Friston
  • Non-parametric FDR as an alternative to Randomize
• Explanation of TFCE (implemented in FSL)

Voodoo correlations and circular inference

Author	Journal	Year	Title
Vul and Pashler	NeuroImage	2012	Voodoo and circularity errors
Kriegeskorte et al.	Nature Neuro	2009	Circular analysis in systems neuroscience: the dangers of double dipping
Kriegeskorte et al.	J Cereb Blood Flow and Metabolism	2010	Everything you never wanted to know about circular analysis, but were afraid to ask

Correlating activation with behaviour

• Lebreton et al, Nature Human Behaviour 2019, Assessing inter-individual differences with task-related functional neuroimaging

ROI analysis

• Guide and tools for ROI analysis

Batch scripting in SPM

• SPM12 Matlab scripting tutorial

Interpreting activations

Interpreting activations with fMRI is not always straightforward, and some caveats should be borne in mind.

Issues with cognitive subtraction

• Friston et al, NeuroImage 1996, The trouble with cognitive subtraction

Reverse inference

• Poldrack, TiCS 2006, Can cognitive processes be inferred from neuroimaging data?
• Hutzler, NeuroImage 2014, Reverse inference is not a fallacy per se: Cognitive processes can be inferred from functional imaging data

Issues with group averages

• Seghier and Price, TiCS 2018, Interpreting and Utilising Intersubject Variability in Brain Function

Resting-state fMRI

General overviews

• Binder, NeuroIma 2012, Task-induced deactivation and the "resting" state
• Biswal, NeuroImag 2012, Resting state fMRI: A personal history
• Lowe, NeuroImage 2012, The emergence of doing “nothing” as a viable paradigm design
• Snyder and Raichle, NeuroImage 2012, A brief history of the resting state: The Washington University perspective

OHBM blog post on considerations when performing a resting-state analysis

Pre-processing

The problem of physiological noise

• Cordes et al, Neuroardiol 2001, Frequencies Contributing to Functional Connectivity in the Cerebral Cortex in ‘‘Resting-state’’ Data
• Birn et al, NeuroImage 2006, Separating respiratory-variation-related fluctuations from neuronal-activity-related fluctuations in fMRI
• Birn et al, HBM 2008, The Effect of Respiration Variations on Independent Component Analysis Results of Resting State Functional Connectivity

A list of physiological processes contributing to the fMRI signal

Global signal correction

• Fox & Raichle, Nature Reviews Neuroscience, 2007, Spontaneous Fluctuations in Brain Activity Observed with Functional Magnetic Resonance Imaging
• Fox et al., Journal of Neurophysiology, 2009, The Global Signal and Observed Anticorrelated Resting State Brain Networks
• Glasser et al., NeuroImage, 2018, Using Temporal ICA to Selectively Remove Global Noise While Preserving Global Signal in Functional MRI Data
• Murphy et al., NeuroImage, 2009, The Impact of Global Signal Regression on Resting State Correlations: Are Anti-Correlated Networks Introduced?
• Murphy & Fox, NeuroImage, 2017, Towards a Consensus Regarding Global Signal Regression for Resting State Functional Connectivity MRI
• Power et al., NeuroImage, 2017, Sources and Implications of Whole-Brain fMRI Signals in Humans
• Power, NeuroImage, 2019, Temporal ICA Has Not Properly Separated Global fMRI Signals: A Comment on Glasser et al. (2018)
• Saad et al., Brain Connectivity, 2012, Trouble at Rest: How Correlation Patterns and Group Differences Become Distorted After Global Signal Regression
• Saad et al., Brain Connectivity, 2013, Correcting Brain-Wide Correlation Differences in Resting-State FMRI
• Zhang & Northoff, Communications Biology, 2022, Beyond Noise to Function: Reframing the Global Brain Activity and Its Dynamic Topography

Data-based physiological correction

• Bezhadi et al, NeuroImage 2007, A component based noise correction method (CompCor) for BOLD and perfusion based fMRI
• Prium et al, NeuroImage 2015, ICA-AROMA
• Salimi-Khorshidi et al, NeuroImage 2014, Automatic denoising of functional MRI data: Combining independent component analysis and hierarchical fusion of classifiers

Head motion

First identification of the problem of micro-movements & commentaries

• Power et al, NeuroImage 2012, Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion
• Carp, NeuroImage 2012, Optimizing the order of operations for movement scrubbing: Comment on Power et al.
• Power et al, NeuroImage 2012, Steps toward optimizing motion artifact removal in functional connectivity MRI; a reply to Carp
• Power et al, NeuroImage 2015, Recent progress and outstanding issues in motion correction in resting state fMRI

Source of head motion in fMRI

Head motion and physiology in multi band fMRI

Good review of issues and corrections

Various comparisons of methods for addressing micro-movement problem

• Satterthwaite et al., NeuroImage, 2013, An Improved Framework for Confound Regression and Filtering for Control of Motion Artifact in the Preprocessing of Resting-State Functional Connectivity Data
• Satterthwaite et al., NeuroImage, 2012, Impact of In-Scanner Head Motion on Multiple Measures of Functional Connectivity: Relevance for Studies of Neurodevelopment in Youth
• Satterthwaite et al., NeuroImage, 2013, Heterogeneous Impact of Motion on Fundamental Patterns of Developmental Changes in Functional Connectivity during Youth
• Yan et al., NeuroImage, 2013, A Comprehensive Assessment of Regional Variation in the Impact of Head Micromovements on Functional Connectomics
• Ciric et al., NeuroImage, 2017, Benchmarking of Participant-Level Confound Regression Strategies for the Control of Motion Artifact in Studies of Functional Connectivity
• Parkes et al., NeuroImage, 2018, An Evaluation of the Efficacy, Reliability, and Sensitivity of Motion Correction Strategies for Resting-State Functional MRI