Dear all,
How can one define visuo-spatial working memory capacity of subjects reporting a continuous value? Subjects are shown a set of stimuli and after a given delay, they have to report using a cursor where one of the stimulus was shown. I'm aware of this kind measure for discrete reports, but I guess those are not useful here.
Thank you in advance,
You can look at the probability of a report from memory times the set size. If you are modelling the response according to any number of mixture models there is typically a value for the proportion of guessing. You can take [(1-guessing)*set size] to estimate the number of items in memory; you are limited by the goodness of your model fit.
Another approach that Wei Ji Ma suggested was reporting the standard deviation of recall errors as a function of set size, and compare the functions. See Ma, Husain, & Bays, 2014 for a review.
Dear João! You as an expert are well aware that in this matter there are different views on the definition and testing methods. I usually guided by the views of various experts (the text of some articles are attached).
Vladimir
You can look at the probability of a report from memory times the set size. If you are modelling the response according to any number of mixture models there is typically a value for the proportion of guessing. You can take [(1-guessing)*set size] to estimate the number of items in memory; you are limited by the goodness of your model fit.
Another approach that Wei Ji Ma suggested was reporting the standard deviation of recall errors as a function of set size, and compare the functions. See Ma, Husain, & Bays, 2014 for a review.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics