Advances in Event-Related fMRI Design Douglas N. Greve

Outline What is Event-Related Design? Blocked Design Fixed-Interval Event-Related Rapid-Presentation (Jittered) Event-Related Efficiency and Event Scheduling Blocked Design gives context for ER. RP-ER brings problem of scheduling.

Fact of (fMRI) Life #1: Dispersion Not like EEG/MEG where response is finished after a few hundred ms. How closely can trials/events be placed and still be able to differentiate their responses? This data is the average of 212 responses. How closely can trials/events be spaced?

Fact of (fMRI) Life #2: Noise Ultimately, noise is the rate limiting factor. The further away from baseline, the easier it is to detect the response. These are stddev bars. Error bars will shrink as more data are collected. To detect a response, need to (1) create a bigger signal, and/or (2) collect many trials. Down side to many trials: (1) cost, (2) wear and tear on subjects, (3) need all that stimuli. How much data needs to be collected?

Fact of (fMRI) Life #3: Time Collect lots of observations to reduce noise Time is Money Subjects won’t work forever Stimuli – famous faces, non-words, syntactically anomalous sentences.

Blocked Design Consecutive, rapid presentation for long duration. fixation Consecutive, rapid presentation for long duration. Use overlap to build a larger signal. Simple analysis. Optimal for detection. Each color represents a different type of stimulus. May be a stimulus or task of long duration instead of multiple rapid presentations. Example of stimulus types: global/local, novel/repeated, happy/sad faces. Assumption: invariant asymptotic. Use long dispersion to create overlap and generate a larger signal.

Using Overlap to Increase Amplitude The effect of rapid presentation on the amplitude of the response. Don’t get 3 times the amplitude with 3 trials. The greater the amplitude, the more it stands out from the noise. 1 2 3

Blocked Design Drawbacks Lose ability to distinguish individual responses Confounding psychological and physiological effects Habituation/Adaptation Expectation Set (Strategy) Experimenter may be trying to measure set. Highly constrained design. Not good for Multi-modal. Little info about hemodynamics. Habituation – that face AGAIN. Expectation/Anticipation. Reminder: efficient.

What is Event-Related Design? (c.f. Blocked Design) Measure Average Response to Single Event Type Post Hoc Event Assignment based on Subject’s Response Random Order of Events Historical: EEG/Evoked Potentials Less Powerful than Blocked Already described an event in first slide. Single trial.

Fixed-Interval Event-Related 12-20s Push trials apart enough to prevent overlap. Interval fixed at minimum is most efficient. Random Sequence (Counter-balanced) Allows Post-Hoc Stimulus Definition Mitigates Habituation, Expectation (?), and Set Inflexible/Inefficient/Boring Good if limited by number of stimuli (not scanning time) Blue bar is length of typical block from blocked design. Raw signal still interpretable. Assumption: time-invariant response. Not very many presentations. Most efficient without overlap is fixed at minimum time needed to prevent overlap.

Rapid-Presentation Event-Related Closely Spaced Trials (Overlap!) Raw signal uninterpretable More Stimulus Presentations for given scanning interval Random Sequence Jitter = “Random” Inter-Stimulus Interval (ISI/SOA) It is possible to present stimuli so close to each other that they overlap and still distinguish between responses. Must observe all levels of overlap. Raw signal uninterpretable. ISI/SOA is a randomly chosen multiple of the TR.

Where does jitter come from? (What’s a Null Condition?) “Null” condition – fixation cross or dot By hypothesis, no response to null Insert random amounts of null between task conditions Differential ISI = Differential Overlap A + B A + A B A + + B + B A The null condition is not a condition. Differential overlap and gender/handedness. Conditions A and B with fixation (+). Yellow bars indicate the ISI between A trials. Red bars indicate ISI between B trials. Green bars indicate ISI from A to B. Differential overlap can come from non-adjacent presentations. Zero ISI. No bars between null and A or B because null is not a condition. Time

Rapid-Presentation Properties Efficient (not as efficient as blocked) Can distinguish responses despite overlap Highly resistant to habituation, set, and expectation Flexible timing (Behavioral, EEG, MEG) Linear overlap assumption Analysis: Selective Averaging/Deconvolution (GLM) How to schedule stimulus onsets?

Scheduling and Efficiency B: N=10 C: N=10 Efficiency: statistical power/SNR/CNR per acquisition Efficiency increases with N Efficiency decreases with overlap Efficiency increases with differential overlap Choose schedule with optimum efficiency before scanning Three designs: A, B, and C all within a fixed scanning interval. A has no overlap but only N=5. B has N=10 but no differential overlap. C has N=10 and differential overlap. Larger the N, the lower the ISI. Differential overlap: cf gender/handedness test.

Summary Facts of Life: Dispersion, Noise, Time Blocked - Habituation, Expectation, Set, No Post-Hoc Fixed-Interval Event-Related – Inefficient/Boring Rapid-Presentation Event-Related Randomized inter-stimulus onsets Overlap Linearity Efficient - Optimization Tool Identical designs for Behavioral, fMRI, EEG, and MEG Blocked uses dispersion to create a bigger signal at the cost of distinguishing individual responses (and hab, exp, and set).