Re: Signal detection: signal, noise and a 2nd signal?
Sounds just perfect. Thanks for taking the time! [EMAIL PROTECTED] wrote: The question of discriminating among three or more events has been successfully tackled by Brian Scurfield. He extended typical two-event ROC analysis to n-event ROC analysis (n2), where results are expressed as n-dimentional ROC hypersurfaces, and sensitivity can be understood in terms of hypervolumes under the hypersurfaces. He also developed a new type of distribution-free sensitivity measure based on an information theory analysis of n-event discrimination tasks. The measure gives an overall measure of detectability among n events, and also allows sensible comparisons to be made between n-event tasks and (n-1)-event tasks, say. Scurfield illustrated his findings using the 3-event case, so if you're specifically interested in that case, check out his papers: Scurfield, B.K. (1996) "Multiple-event forced-choice tasks in the theory of signal detectability", Journal of Mathematical Psychology, 40(3), 253-269 Scurfield, B.K. (1998) "Generalization of the theory of signal detectability to m-dimensional n-event forced-choice tasks", Journal of Mathematical Psychology, 42(1), 5-31. The JMP abstracts used to be available online, but I don't know if they still are. Also, there was an independent development of some of this material by Douglas Mossman. He had a paper in Medical Decision Making in either 1998 or 1999 entitled "Three-way ROCs". Sorry, can't remember the volume. Hope this helps, Vit D. -- Manni === This list is open to everyone. Occasionally, less thoughtful people send inappropriate messages. Please DO NOT COMPLAIN TO THE POSTMASTER about these messages because the postmaster has no way of controlling them, and excessive complaints will result in termination of the list. For information about this list, including information about the problem of inappropriate messages and information about how to unsubscribe, please see the web page at http://jse.stat.ncsu.edu/ ===
Signal detection: signal, noise and a 2nd signal?
Hi! We are doing research in visual perception. To measure subjects ability two perceive certain stimuli we computed d' or ds. But these measures only provide information about the ability to detect one signal. We are in the process of designing a new experiment and we would like to know, whether subjects can discriminate 3 different stimuli. Unfortunately we do not see a way to use signal detection theory here. The problem is, that you can tell a hit from some error, but what would be a false alarm? Does anyone know of a statistical procedure that would allow us to compute one or several indices, that could be used to measure subjects performance? Of course this would also influence the question(s) we ask in the experiments and probably the number of trials we need per subject and stimulus. But since we are still in the design stage, this would not be a great difficulty. Thanks, -- Manni === This list is open to everyone. Occasionally, less thoughtful people send inappropriate messages. Please DO NOT COMPLAIN TO THE POSTMASTER about these messages because the postmaster has no way of controlling them, and excessive complaints will result in termination of the list. For information about this list, including information about the problem of inappropriate messages and information about how to unsubscribe, please see the web page at http://jse.stat.ncsu.edu/ ===
Re: Signal detection: signal, noise and a 2nd signal?
[EMAIL PROTECTED] (Donald F. Burrill) wrote: On Sat, 20 May 2000, Manni Heumann wrote: We are doing research in visual perception. To measure subjects ability two perceive certain stimuli we computed d' or ds. But these measures only provide information about the ability to detect one signal. I don't know what d' or ds are (or perhaps I do not recognize them in this rather spare context); so it's difficult to guess whether there simply do not exist analogues of them in multiple-signal contexts, or there are different ways of defining analogues and you haven't managed to settle on a suitable way. d' and ds are signal detection theory measures of sensitivity. Unfortunately, I don't know of any analogues, but I guess, that there aren't. We are in the process of designing a new experiment and we would like to know, whether subjects can discriminate 3 different stimuli. Do you really want to know "whether" (for which the answer must surely be "Yes", at some (possibly uninteresting) level of discourse)? Or do you want to know "how well" or "how reliably" the stimuli can be discriminated? We expect, that subjects cannot discriminate between the stimuli, or at least that is what we hope for. So a dichotomous measure would be enough, although it should be rather sensitive. Unfortunately we do not see a way to use signal detection theory here. The problem is, that you can tell a hit from some error, but what would be a false alarm? Another kind of error, presumably. With only 1 stimulus of interest, you have two kinds of "hits" (yes, the stimulus is present; no, it isn't) and two kinds of "misses" or errors (no it isn't, but in fact it was; yes it is but in fact it wasn't, or a "false alarm"). With 3 stimuli of interest you have more kinds of "hits" and even more kinds of "misses". The false alarms are critical. The advantage of d' or ds is that they are not affected by response bias. Suppose a subject answered the question, whether the signal was present, consistently with "Yes!". That would result in 100% hits. On the other hand you'd also get 100% false alarms. Simplifying, you can say that d'=hits-falseAlarms. To me the distinction of a false alarm and a miss is vital, but seems to break down in the multi signal context. -- Manni === This list is open to everyone. Occasionally, less thoughtful people send inappropriate messages. Please DO NOT COMPLAIN TO THE POSTMASTER about these messages because the postmaster has no way of controlling them, and excessive complaints will result in termination of the list. For information about this list, including information about the problem of inappropriate messages and information about how to unsubscribe, please see the web page at http://jse.stat.ncsu.edu/ ===
Newbie Question
Hi! I just discovered the newsgroups sci.stat.consult, sci.stat.edu, and sci.stat.math, and I was wondering, which group covers which topics. Are there any FAQs for any of these groups? What is on, what is off topic? TIA, -- Manni === This list is open to everyone. Occasionally, less thoughtful people send inappropriate messages. Please DO NOT COMPLAIN TO THE POSTMASTER about these messages because the postmaster has no way of controlling them, and excessive complaints will result in termination of the list. For information about this list, including information about the problem of inappropriate messages and information about how to unsubscribe, please see the web page at http://jse.stat.ncsu.edu/ ===