The Effect of Stress on Cognitive Load Measurement

Conway, Dan; Dick, Ian; Li, Zhidong; Wang, Yang; Chen, Fang

doi:10.1007/978-3-642-40498-6_58

Cited by 29 publications

(21 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The more numbers students had to remember and the more mathematical problems they had to solve, the harder it was for students to complete the test correctly which might have induced both high CL and stress. This would be in accordance with the literature since task failure and feelings of lack of control have been shown to induce stress (Conway, Dick, Li, Wang, & Chen, 2013).…”

Section: Main Findings and Implicationssupporting

confidence: 92%

Multimodal learning analytics to investigate cognitive load during online problem solving

Larmuseau¹,

Cornelis²,

Lancieri³

et al. 2020

Brit J Educational Tech

View full text Add to dashboard Cite

To have insight into cognitive load (CL) during online complex problem solving, this study aimed at measuring CL through physiological data. This study experimentally manipulated intrinsic and extraneous load of exercises in the domain of statistics, resulting in four conditions: high complex with hints, low complex with hints, high complex without hints and low complex without hints. The study had a within‐subject‐design in which 67 students solved the exercises in a randomized order. Self‐reported CL was combined with physiological data, namely, galvanic skin response (GSR), skin temperature (ST), heart rate (HR) and heart rate variability (HRV). Multiple imputation was used for handling missing data from resp. 16 and 19 students for GSR/ST and HR/HRV. First, differences between conditions in view of physiological data were examined. Second, we investigated how much variance of self‐reported CL and task performance was explained by physiological data. Finally, we investigated which features can be used to assess (objective) CL. Results revealed no significant differences between the manipulated conditions in terms of physiological data. Nonetheless, HR and ST were significantly related to self‐reported CL, whereas ST to task performance. Additionally, this study revealed the potential of ST and HR to assess high CL.

show abstract

Section: Main Findings and Implicationssupporting

confidence: 92%

Multimodal learning analytics to investigate cognitive load during online problem solving

Larmuseau¹,

Cornelis²,

Lancieri³

et al. 2020

Brit J Educational Tech

View full text Add to dashboard Cite

show abstract

“…HRV features can be an important indicator for classifying cognitive load because cognitive load modulates the sympathetic and parasympathetic nervous systems inversely to driver sleepiness [98]. The time domain GSR, i.e., the peak amplitude, the duration of the rise time of each peak, and the mean GSR value were found to be useful indicators for cognitive load detection when a person is under the influence of different stress levels [99]. Furthermore, the states depend on the experimental design, driving environment, confounding factors, etc., and hence, multi-variate data and data fusion considering the driving context are needed to accurately assess the cognitive load.…”

Section: Discussionmentioning

confidence: 99%

Towards Intelligent Data Analytics: A Case Study in Driver Cognitive Load Classification

Ahmed

Begum

2020

Brain Sciences

View full text Add to dashboard Cite

One debatable issue in traffic safety research is that the cognitive load by secondary tasks reduces primary task performance, i.e., driving. In this paper, the study adopted a version of the n-back task as a cognitively loading secondary task on the primary task, i.e., driving; where drivers drove in three different simulated driving scenarios. This paper has taken a multimodal approach to perform ‘intelligent multivariate data analytics’ based on machine learning (ML). Here, the k-nearest neighbour (k-NN), support vector machine (SVM), and random forest (RF) are used for driver cognitive load classification. Moreover, physiological measures have proven to be sophisticated in cognitive load identification, yet it suffers from confounding factors and noise. Therefore, this work uses multi-component signals, i.e., physiological measures and vehicular features to overcome that problem. Both multiclass and binary classifications have been performed to distinguish normal driving from cognitive load tasks. To identify the optimal feature set, two feature selection algorithms, i.e., sequential forward floating selection (SFFS) and random forest have been applied where out of 323 features, a subset of 42 features has been selected as the best feature subset. For the classification, RF has shown better performance with F1-score of 0.75 and 0.80 than two other algorithms. Moreover, the result shows that using multicomponent features classifiers could classify better than using features from a single source.

show abstract

“…Asking the participant to finish a particular task within a limited time is considered likely to result in stress in the participant, since stress is commonly linked to high levels of cognitive load [11]. Time constraints can be a convenient means, applicable to any task, to further increase the difficulty and induce extreme cognitive load.…”

Section: Time Constraintmentioning

confidence: 99%

Speech Signal Based Measures

Chen

Zhou

Wang

et al. 2016

Robust Multimodal Cognitive Load Measurement

Self Cite

View full text Add to dashboard Cite

Speech is an ideal modality for cognitive load measurement, due to its universal availability, relatively low cost to acquire, low processing requirement for devices and can be largely non-intrusive. A speech-based cognitive load measurement system might potentially be able to continuously monitor the cognitive load of a person without interfering with the task being conducted. One example of such a system that monitors cognitive load in a non-intrusive way, designed specifically for call centers, is a product called BrainGauge.Many speech parameters have been found to be affected by cognitive load, including prosody, spectrum and vocal tract related parameters. For example, Berthold's investigation has shown that features such as the number of sentence fragments (eg whole sentences versus incomplete sentences), self-interruptions and occurrence of self-repairs when speaking, as well as articulation rate were reflective of cognitive load variations [1]. Similar research conducted by Mueller also confirmed that speaking disfluencies, articulation rate, utterance content quality, complexity of sentence construction as well as silence and filled pauses in an utterance could be used to detect cognitive load via speech [2]. Accordingly, various classification methods have been applied to cognitive load examination. Typical methods include Gaussian Mixture Models (GMM), linear kernel Support Vector Machine (SVM) and hybrid SVM-GMMs which take the output of a GMM as the input for a SVM. Channel and speaker normalization is also widely adopted, together with different delta techniques. Furthermore, a Universal Background Model (UBM) is often necessary to reduce the impact of insufficient data available.This chapter will review the recent investigations into cognitive load measurement via speech, which mainly involve:• Experiments utilized to induce different levels of cognitive load; • The various speech features extracted for cognitive load investigation; • The comparison of different classification methods.

show abstract

The Effect of Stress on Cognitive Load Measurement

Cited by 29 publications

References 16 publications

Multimodal learning analytics to investigate cognitive load during online problem solving

Multimodal learning analytics to investigate cognitive load during online problem solving

Towards Intelligent Data Analytics: A Case Study in Driver Cognitive Load Classification

Speech Signal Based Measures

Contact Info

Product

Resources

About