Automatic Speech Feature Extraction for Cognitive Load Classification

Performance of attention-demanding tasks is challenged if motor (e.g., walking) and cognitive (e.g., talking) tasks are carried out simultaneously. These dual-task paradigms have received increasing interest in probing the attentional influence associated with impairments to these systems [1]. For example, gait instabilities in Alzheimer's patients has been suggested to result from impaired attentional faculties impacting balance control [2]. Parkinson's patients have demonstrated inappropriate prioritization during dual-tasking, potentially leading to a higher risk of falls [3]. Furthermore, the controversy surrounding cell phone conversations on driving performance [4] further motivates the need to measure and understand the influence of attentional load.