Take a look at Figure 1 in the output from the model you just ran.
In this figure we have plotted the prior probability distribution
for q on each of trials 1-5. - Hopefully this will be familiar from the coin-tossing example.
The slot machine pictured to the left of the prior for each trial is the outcome from the previous trial,
which is why this is ''?'' on trial 1
- Why are we plotting the prior on each trial here, as opposed to the posterior as we did in Part 1?
- What do you think the dotted red line indicates?
Now take a look at Figure 2.
In this figure we have plotted the model's estimate of (q) across trials (red trace) and also the true
probability that orange is rewarded.
If you want to see the whole probability distribution over q, have a look at Figure 3.
The shading in the background represents the probability distribution on each trial -
like the distributions in Figure 1
The dashed line is the best estimate of q, and the solid white line is the true value of q
as in Figure 2.