Dynamic Causal Models Will Penny Olivier David, Karl Friston, Lee Harrison, Stefan Kiebel, Andrea Mechelli, Klaas Stephan MultiModal Brain Imaging, Copenhagen, October 25-26, 2005 V1V5SPC V1 V5 SPC Wellcome Department of Imaging Neuroscience, ION, UCL, UK.
Contents Neurodynamic model Hemodynamic model Model estimation and comparison Attention to visual motion Friston et al.(2003) Neuro- Image, 19 (4), pp
Contents Neurodynamic model Hemodynamic model Model estimation and comparison Attention to visual motion
Single region u2u2 u1u1 z1z1 z2z2 z1z1 u1u1 a 11 c
Multiple regions u2u2 u1u1 z1z1 z2z2 z1z1 z2z2 u1u1 a 11 a 22 c a 21
Modulatory inputs u2u2 u1u1 z1z1 z2z2 u2u2 z1z1 z2z2 u1u1 a 11 a 22 c a 21 b 21
Reciprocal connections u2u2 u1u1 z1z1 z2z2 u2u2 z1z1 z2z2 u1u1 a 11 a 22 c a 12 a 21 b 21
Neurodynamics Inputs Change in Neuronal Activity Neuronal Activity Intrinsic Connectivity Matrix Modulatory Connectivity Matrices Input Connectivity Matrix V1 V5 SPC
Contents Neurodynamic model Hemodynamic model Model estimation and comparison Attention to visual motion Single word processing
Hemodynamics Hemodynamic variables For each region: Hemodynamic parameters Seconds Dynamics
Why have explicit models for neurodynamics and hemodynamics ? For 4 event types u 1, u 2, u 3, u 4 : In a GLM for a single region, y=X +e, with 3 basis functions per event type (canonical,shifter, stretcher) there are 12 parameters to estimate. These relate hemodynamics directly to each stimulus. In a (single region) DCM there are 4 neuronal efficacy parameters relating neuronal activity to each stimulus And 5 hemodynamic parameters relating neuronal activity to the BOLD signal. A total of 9 parameters.
DCM Priors Hemodynamics Rate of signal decay: 0.65 Elimination rate: 0.41 Transit time: 0.98 Grubbs exponent: 0.32 Oxygenation fraction: 0.34 E[h] Cov[h] Neurodynamics Stability priors ensure principal Lyapunov exponent is less than zero with high probability.
Contents Neurodynamic model Hemodynamic model Model estimation and comparison Attention to visual motion Single word processing
Bayesian Estimation Relative Precision Weighting Normal densities
Multiple parameters One-step if C e, C p and p are known General Linear Model
Nonlinear models Gauss-Newton ascent with priors Linearization Current Estimates Friston et al.(2002) Neuro- Image, 16 (2), pp
Model Comparison I V1 V5 SPC Model, m Parameters: Prior Posterior Likelihood Evidence Laplace, AIC, BIC approximations Model fit + complexity Penny et al. (2004) NeuroImage, 22 (3), pp
Model Comparison II V1 V5 SPC Model, m Parameters: Prior Posterior Likelihood Prior Posterior Evidence Parameter Model
Model Comparison III V1 V5 SPC Model, m=i V1 V5 SPC Model, m=j Model Evidences: Bayes factor: 1 to 3: Weak 3 to 20: Positive 20 to 100: Strong >100: Very Strong
Contents Neurodynamic model Hemodynamic model Bayesian estimation Attention to visual motion Single word processing
Attention to Visual Motion STIMULI 250 radially moving dots at 4.7 degrees/s PRE-SCANNING 5 x 30s trials with 5 speed changes (reducing to 1%) Task - detect change in radial velocity SCANNING (no speed changes) 6 normal subjects, scan sessions; each session comprising 10 scans of 4 different condition 1.Photic 2.Motion 3.Attention Experimental Factors Buchel et al. 1997
Specify regions of interest Identify regions of Interest eg. V1, V5, SPC GLM analysis V1 V5 SPC Motion Photic Att Model 1
V1 V5 SPC Motion Photic Att Model 1 V1 V5 Estimation SPC Time (seconds)
Posterior Inference B 3 21 P(B 3 21 |y) How much attention (input 3) changes connection from V1 (region 1) to V5 (region(2)
V1 V5 SPC Motion Photic Att Model 1 Motion Photic Att V1 V5 SPC Model 2 Bayes Factor B 12 > Very Strong
V1 V5 SPC Motion Photic Att Model 1 V1 V5 SPC Motion Photic Att Model 3 Bayes Factor B 13 =3.6 Positive
V1 V5 SPC Motion Photic Att Model 1 Motion Photic Att V1 V5 SPC Model 4 Bayes Factor B 14 =2.8 Weak Penny et al. (2004) NeuroImage, Special Issue.
Summary Neurodynamic model Hemodynamic model Bayesian estimation Attention to visual motion