Table 2 Results of the Bayesian model comparison of the Bayesian Causal Inference, the forced-fusion and the full-segregation model

From: The neural dynamics of hierarchical Bayesian causal inference in multisensory perception

 

p common

µ P

σ P

σ A

σ V

R 2

relBIC

pEP

% win

Causal Inference (model averaging)

0.42 ± 0.05

2.26 ± 0.20

2.34 ± 0.29

0.53 ± 0.03

1.11 ± 0.23

0.874 ± 0.012

0

1

95.7

Forced fusion

-

2.10 ± 0.22

4.38 ± 0.60

1.17 ± 0.04

1.45 ± 0.04

0.617 ± 0.016

8362.42

0

4.3

Full segregation

-

2.15 ± 0.20

2.11 ± 0.32

0.55 ± 0.03

1.01 ± 0.11

0.846 ± 0.015

920.70

0

0

  1. Note: pcommon, causal prior; µP, mean of the numeric prior; σP, standard deviation of the numeric prior; σA, standard deviation of the auditory likelihood; σV, standard deviation of the visual likelihood; R2, coefficient of determination; relBIC, Bayesian information criterion at the group level, i.e. participant-specific BICs summed over all participants (BIC = LL − 0.5 m ln(n), LL = log likelihood, m = number of parameters, n = number of data points) of a model relative to the Bayesian Causal Inference (“model averaging”) model (n.b. a smaller relBIC indicates that a model provides a better explanation of our data); pEP, protected exceedance probability, i.e. the probability that a given model is more likely than any other model, beyond differences due to chance). % win, percentage of participants in which a model won the within-participant model comparison based on BIC
Back to article page