Formant Frequencies under Cognitive Load: Effects and Classification

Tet Fei Yap,Julien Epps,Eric H C Choi (Eurasip Member),Eliathamby Ambikairajah

doi:10.1155/2011/219253

Abstract

Cognitive load measurement systems measure the mental demand experienced by human while performing a cognitive task, which is useful in monitoring and enhancing task performance. Various speech-based systems have been proposed for cognitive load classification, but the effect of cognitive load on the speech production system is still not well understood. In this work, we study formant frequencies under different load conditions and utilize formant frequency-based features for automatic cognitive load classification. We find that the slope, dispersion, and duration of vowel formant trajectories exhibit changes under different load conditions; slope and duration are found to be useful features in vowel-based classification. Additionally, 2-class and 3-class utterance-based classification results, evaluated on two different databases, show that the performance of frame-based formant features was comparable, if not better than, baseline MFCC features.

Highlights

Cognitive load refers to the load imposed by a certain task on the cognitive system of a person [1]
The classification results are compared with mel-frequency cepstral coefficients (MFCCs) features (7 MFCC, not including the zeroth coefficient), which are commonly used as a baseline in previous cognitive load classification systems [9, 13, 28]
The first thing to notice is that the first three formant frequencies {F1, F2, F3} outperformed MFCC in both the 2-class and 3-class classification results. This is remarkable given that the formant features have lower dimensionality compared with MFCC (3 and 7, resp.), and the fact that the formant frequencies were not manually corrected in this experiment

Summary

Introduction

Cognitive load refers to the load imposed by a certain task on the cognitive system of a person [1]. First developed in the field of educational psychology, was originally focused on improving the process of acquiring and applying new knowledge [2]. Central to cognitive load theory is the notion that working memory is required while performing a cognitive task, but this working memory is limited. As a cognitive task becomes more challenging, the amount of working memory required to complete the task will typically increase. When the working memory requirement exceeds the available capacity, task performance will deteriorate. There has been a growing interest in monitoring and measuring cognitive load as a means to monitor or even enhance human task performance. Applications expected to benefit from such cognitive load monitoring systems include air traffic control systems [3], in-car user interfaces [4] and military human-machine systems [5]

Methods

Results

Conclusion