Subject: Re: identification task...negative d' values. From: Massimo Grassi <massimo.grassi@xxxxxxxx> Date: Wed, 3 May 2006 09:34:57 +0200Dear ... (what's your name?) > I have some few small negative values of d' obtained from an > identification task in witch subjects label vowel sounds from a > continuum with tree phoneme labels. Generally, the graphs and tables > in the literature never show negative values of d' ( i think it > occurs here and there). how are you calculating d'? If you use excel the formula is the following: d' = normsinv(H) - normsinv(FA) where H and FA are, respectively, Hits and False Alarms probabilities. If you were simply inverting H and FA... However, I don't understand one thing. You write "subjects label vowel sounds from a continuum with tree phoneme labels". So you present me a sound that is between /a/ and /i/ and I have to decide whether I perceive /a/ or /i/. Is d' the appropriate analysis? Ciao, m ******************** Massimo Grassi - PhD Laboratorio di Psicologia Via Petracco 8 - 33100 Udine - Italy http://www.psy.unipd.it/~grassi IMPORTANT! BEFORE SENDING REGULAR MAIL PLEASE SEND AN EMAIL ---------------------------------------------------------------------- SEMEL (SErvizio di Messaging ELettronico) - CSIT -Universita' di Udine