score = n_level + (percent_correct -
threshold_fallback)/(threshold_advance - threshold_fallback)
so:
n, % | score
-------------------
4, 0% | 2.33
4, 10% | 2.66
4, 20% | 3.00
4, 30% | 3.33
4, 40% | 3.66
4, 50% | 4.00
4, 60% | 4.33
4, 70% | 4.66
4, 80% | 5.00
4, 90% | 5.33
4, 100% | 5.66
I made the scoring work like this because I thought it gave a better
representation of your actual performance than just the average n-back
level, especially when you have a hard time getting past a level. I
really ought to document it better, though, or change the text labels
for the graph.
Undocumented hack feature: press 'm' at the graphing screen to play
around with different scoring formulae. Good luck guessing what is what,
though.
Jonathan Toomim
--
You received this message because you are subscribed to the Google Groups "Dual N-Back, Brain Training & Intelligence" group.
To post to this group, send email to brain-training@googlegroups.com.
To unsubscribe from this group, send email to brain-training+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/brain-training?hl=en.