Date：2024/11/27 (Wed)
Venue：N100, Department of Psychology
14:30-15:00
Speaker：Feng-Chun Chou
Title：Decoding Preferences: Neural Mechanisms Linking Appraisal to Choice
Abstract：
Making decisions is an essential part of our daily lives, guided by how we perceive and appraise external events or stimuli. Interestingly, even when individuals share similar appraisals, they often make different choices based on their personal preferences. This study aims to examine the underlying neural mechanisms connecting appraisal and decision-making. Seventy participants who passively viewed short video clips with diverse topics underwent functional scanning within the fMRI scanner without any prior instruction, ensuring that their appraisals of these videos were not biased by any prior expectation. Behavioral appraisal models utilized behavioral appraisal features, such as emotional engagement and narrative coherence ratings, and multivariate predictive brain models were trained individually for each participant. A 5-fold cross-validation method was employed to validate the performance of these models, which performed significantly above chance. Intersubject representational similarity analysis further revealed a strong association between intersubject similarities in behavioral appraisal models and similarities in multivariate brain models, highlighting the complicated nature of decision-making processes. These findings deepen our understanding of the neural mechanisms underlying preference formation.
15:00-15:30
Speaker：Hau-Hung Yang
Title：On the application of stochastic approximation to computerized adaptive testing
Abstract：
In the study of item response theory (IRT), the maximum information (item selection) method (or procedure, rule) is prevailing in test constructions, including the computerized adaptive testing (CAT). However, this method may not be suitable if the trial number is small in the CAT. In this study, we advocate the use of the stochastic-approximation based rule for item difficulty determination for short test lengths in the CAT. We also describe a generalized stochastic-approximation rule to take item discrimination into account. In a simulation study, we considered two cases of the IRT, namely the Rasch model and the 2PL model, and for each case compared the performance of the information-based and stochastic-approximation based procedures for trials from 10 to 60. The results showed that the accelerated stochastic approximation procedure (and its generalization) was more efficient than the information-based method across the trials. Further, in both procedures the bias of the estimator started to diminish quickly after the early stages of trials.