Effective Connectivity Analysis of the Brain Network in Drivers during Actual Driving Using Near-Infrared Spectroscopy.

Zhian Liu,Congcong Huo,Zengyong Li,Qitao Tan,Quan Yuan,Gongcheng Xu,Ming Zhang

doi:10.3389/fnbeh.2017.00211

Abstract

Driving a vehicle is a complex activity that requires high-level brain functions. This study aimed to assess the change in effective connectivity (EC) between the prefrontal cortex (PFC), motor-related areas (MA) and vision-related areas (VA) in the brain network among the resting, simple-driving and car-following states. Twelve young male right-handed adults were recruited to participate in an actual driving experiment. The brain delta [HbO2] signals were continuously recorded using functional near infrared spectroscopy (fNIRS) instruments. The conditional Granger causality (GC) analysis, which is a data-driven method that can explore the causal interactions among different brain areas, was performed to evaluate the EC. The results demonstrated that the hemodynamic activity level of the brain increased with an increase in the cognitive workload. The connection strength among PFC, MA and VA increased from the resting state to the simple-driving state, whereas the connection strength relatively decreased during the car-following task. The PFC in EC appeared as the causal target, while the MA and VA appeared as the causal sources. However, l-MA turned into causal targets with the subtask of car-following. These findings indicate that the hemodynamic activity level of the cerebral cortex increases linearly with increasing cognitive workload. The EC of the brain network can be strengthened by a cognitive workload, but also can be weakened by a superfluous cognitive workload such as driving with subtasks.

Highlights

Driving a vehicle is a complex activity that requires high-level brain functions, such as planning, decision making, visual attention, motor control and high cognitive activity to make fast cognitive decisions in a complex and rapidly changing environment (Derosière et al, 2014)
The wavelet amplitude (WA) of left motor-related areas (MA) and vision-related areas (VA) was higher than that of the right brain areas, and the WA of the bilateral prefrontal cortex (PFC) changed statistically significantly along with the change in the cognitive workload; (2) the connection strength of effective connectivity (EC) increased in the task_1 state as compared with that in the resting state, whereas it decreased in the task_2 state as compared with that in the task_1 state; and (3) the bilateral PFC were causal targets, and the bilateral VA and MA were causal sources except for l-MA became a causal target in the task_2 state. These findings suggest that the EC of the brain network can be strengthened by a cognitive workload, and can be weakened by a superfluous cognitive workload such as driving with a car-following task
The connection strength of EC was enhanced from the resting state to the simpledriving state, but it deteriorated during the car-following task

Summary

Introduction

Driving a vehicle is a complex activity that requires high-level brain functions, such as planning, decision making, visual attention, motor control and high cognitive activity to make fast cognitive decisions in a complex and rapidly changing environment (Derosière et al, 2014). It makes sense to analyze driving data in a manner that evaluates the brain activation among regions, which enables us to study how the brain is functionally connected and how these intrinsic networks are modulated by the cognitive workload (Calhoun and Pearlson, 2012). The authors found that the signal in frontoparietal regions decreased exponentially with a rate proportional to the driving speed. They found that increases in the cerebella and occipital areas, presumably related to the complex visuomotor integration, were activated during driving but not associated with the driving speed (Calhoun et al, 2002)

Objectives

Methods

Results

Discussion

Conclusion