Table 3 Results of the Bayesian softmax-regression model for subjective expectation/experience.

From: Placebo Intervention Enhances Reward Learning in Healthy Individuals

Parameter

Expected/Experienced effect

Posterior distribution

Mean a

sd b

95% HDI c

intercept

negative

−1.72

0.87

[−3.49, −0.11]

 

positive

0.60

0.43

[−0.23, 1.45]

condition

negative

0.50

1.13

[−1.66, 2.77]

 

positive

−0.06

0.61

[−1.25, 1.15]

question

negative

−1.53

1.57

[−4.72, 1.50]

 

positive

−0.92

0.59

[−2.10, 0.21]

condition × question

negative

2.02

1.79

[−1.41, 5.64]

 

positive

−0.51

0.89

[−2.25, 1.24]

  1. Coefficients denote changes in log-odds relative to “neutral”. There is no apparent effect of the placebo intervention on anticipated or experienced performance.
  2. aMean of the posterior distribution marginalized for each variable.
  3. bStandard deviation of the marginal posterior distribution
  4. cHDI = highest density interval.