|
Signal Detection Theory provides a valuable theoretical model to investigate mechanisms of human perception. On one side, it accounts for the contribution of noise, both exogenous and endogenous, by assuming normal distributed internal representation of stimuli. On the other side, because of the dichotomic nature of the subject response, the existence of an internal criterion is required; such a criterion can be in some way modulated by the subject, tipically in a task aiming to reach the maxiumum percentage of correct responses.
We focus our research on the criterion setting issue with particular attetion to dynamics aspects. By means of feedback techniques, we induce decisional biases in orientation "Yes-No" experiments within the visual modality, so as to investigate the time-domain dynamics of criterion shifts under feedback control. We perform data analysis by fitting an exponential model, which is characteristic of closed-loop systems, to the data, by using maximum likelyhood exstimation techniques on a trial-by-trial basis. |
|