Increased Functional Connectivity Between the Parietal and Occipital Modules Among Flight Cadets
INTRODUCTION: Modular organization in brain regions often performs specific biological functions and is largely based on anatomically and/or functionally related brain areas. The current study aimed to explore changes in whole-brain modular organization affected by flight training. METHODS: The study included 25 male flight cadets and 24 male controls. The first assessment was performed in 2019, when the subjects were university freshmen. The second assessment was completed in 2022. High spatial resolution structural imaging (T1) and resting-state functional MRI data were collected. Then, 90 cerebral regions were organized into 6 brain modules. The intensity of intra- and intermodular communication was calculated. RESULTS: Mixed-effect regression model analysis identified significantly increased interconnections between the parietal and occipital modules in the cadet group, but significantly decreased interconnections in the control group. This change was largely attributed to flight training. DISCUSSION: Pilots need to control the aircraft (e.g., attitude, heading, etc.) using the stick and pedal in response to the current state of the aircraft displayed by the instrument panel; as such, flying requires a large amount of hand–eye coordination. Day-to-day flight training appeared to intensify the connection between the parietal and occipital modules among cadets. Chen X, Jiang H, Meng Y, Xu Z, Luo C. Increased functional connectivity between the parietal and occipital modules among flight cadets. Aerosp Med Hum Perform. 2024; 95(7):375–380.
Commercial aviation is one of the largest industries worldwide. Previously, pilots were mostly recruited from the military, and “stick-and-rudder skills” were the most important abilities. Increasing automation emphasizes nontechnical pilot skills and has made flying more complex. Safe flight operations require dedicated professionalism and a significant amount of specialized training. The Civil Aviation Flight University of China (CAFUC) is the largest university of civil aviation in Asia and has trained >10,000 pilots for civil aviation. Currently, the regulations for airline pilot training are largely based on evidence from accidents involving jet aircraft in earlier generations. During the first 2 yr of study at the CAFUC, cadets are required to complete theoretical aviation training. Subsequently, they complete flight training with approximately 250 h (flight simulator and real aircraft) to learn piloting skills. This flight training has been the standard for many years; however, the effects of flight training on the brain have rarely been studied.
Our previous studies8,10 with pilots had identified the psychophysiological changes of flying. In addition, in a previous study, we explored the local characteristics of brain connectivity networks and identified alterations caused by flight training.9 In the current study, we aimed to explore whole-brain changes caused by flight training. The modularity of the brain presents a balance between local specialization and global integration levels of brain function. It is accepted as one of the central organizational principles of the brain.6,8 Modular organization in brain regions often performs specific biological functions and is largely based on anatomically and/or functionally related brain areas. We collected brain MRI data from flight cadets and controls at two visits. Changes in the modularity of the two groups at two time points were compared. We hypothesized that modularity of the brain may be altered as a result of flight training.
METHODS
Subjects
The present study included 25 male flight cadets and 24 male controls, all of whom were undergraduates enrolled at CAFUC. Subjects were recruited through advertisements and received compensation. Individuals with brain injury, a history of neurological illness, and substance-related disorders were excluded. The two groups were matched for handedness and age. All subjects were naïve to MRI and provided informed written consent to participate in this study.
All cadets completed 40 h of simulation training, with an average actual flight training time of approximately 200 h (Table I). Aircraft types included the C172R and DA-42. The training mainly included the learning of flying skills, but also included the learning of crew resource management, such as interpersonal communication, stress management, decision-making, etc. The control group majored in psychology or aeronautical engineering. None of the controls had any experience with flying or flight simulators.
The study consisted of two assessments, the first of which occurred in 2019, when the subjects were university freshmen (i.e., first-year undergraduates). The second assessment was completed in 2022, with an interval ranging from 31–41 mo across the subjects between visits. During the two visits, high-spatial-resolution structural images (T1) and resting-state functional MRI data were collected.
All data were collected at the Center for Information in Medicine at the University of Electronic Science and Technology of China. The Ethics Committee of the University of Electronic Science and Technology of China (Chengdu, China) approved the study protocol (No. 2019-042,019).
Equipment
All images were acquired using a 3 Tesla MRI scanner (Discovery MR 750; GE Healthcare, Waukesha, WI, USA). A 3-D spoiled gradient echo pulse sequence was used to collect T1 data. The parameters were as follows: repetition time, 5.976 ms; echo time, 1.976 ms; flip angle, 9°; field of view, 256 × 256 × 154 mm; matrix, 256 × 256; slice number, 154; and voxel size, 1 × 1 × 1 mm. A gradient echo planar imaging sequence was used to collect functional data. During scanning, the participants were told to lie still with their eyes closed and not fall asleep. The parameters were as follows: repetition time, 2000 ms; echo time, 30 ms; flip angle, 90°; field of view, 240 mm × 240 mm × 140 mm; matrix, 64 × 64; and slice thickness, 4 mm (no gap). Thus, the in-plane voxel size was 3.75 mm × 3.75 mm × 4 mm. Each volume consisted of 35 slices. In total, 255 volumes were acquired from each subject. The flight training time for each cadet was also collected.
Procedure
MRI data preprocessing was performed using DPARSF_V5.2 (http://rfmri.org/DPARSF). T1 images were segmented into gray matter, white matter, and cerebrospinal fluid. The first 10 scans of functional MRI data were removed to achieve magnetization equilibrium. The remaining scans were subjected to slice timing and head motion correction. Subjects who exceeded a head motion of 2 mm in any direction or 2° rotation in any direction were excluded from the subsequent analysis. The images were then co-registered with the structural gray matter images and normalized to the standard MNI template at 3 × 3 × 3 mm3 resolution. The images were then processed to remove the linear trends and bandpass filtered (0.01–0.08 Hz). White matter, cerebrospinal fluid, and head-motion parameters (rigid-body, 6 parameters) were regressed as nuisance covariates.
GRETNA toolbox software was used to perform subsequent calculations. The preprocessed functional MRI datasets were segmented into 116 regions using an anatomically labeled template provided in the toolbox.33 There were 90 regions in the cerebrum and 26 in the cerebellum. The 90 cerebral regions were organized into 6 brain modules (the frontal, prefrontal, subcortical, parietal, temporal, and occipital modules) (Fig. 1). Detailed information regarding the modules is summarized in the ancillary table (summarized in Table SI in Appendix A; found online at https://doi.org/10.3357/amhp.6370sd.2024).
Citation: Aerospace Medicine and Human Performance 95, 7; 10.3357/AMHP.6370.2024
The mean time series of each of the 90 regions was obtained to calculate Pearson correlation coefficients between each pair of regions. Subsequently, a 90 × 90 functional connectivity matrix was generated for each subject. Only connections with positive values were reserved. Subsequently, a threshold is used to convert the networks into binary networks. When the threshold value of the functional correlation coefficient was 0.36, the small-world properties of the control group were significantly >1.1. Thus, a series of threshold values (0.05:0.01:0.36) was used to obtain reliable results. The sum of the number of significant edges within and between the modules was used to represent the intensity of within-modular and intermodular communication. The area under the curve (AUC) represents the connection values for all ranges of threshold values. The AUC was used for subsequent statistical calculations.
Statistical analysis
A mixed-effects regression model12 was used to compare changes in the AUC within and between the modules of the two groups at two time points (2 * 2 ANOVA analysis, multiple contrast corrected with family-wise error [FWE] correction, P < 0.05). Handedness and mean FD power were used as covariates. The interaction effect between group and time represented the flight training effect. The main effect of the group represented individual differences between the two groups. The main effect of time represented the natural alterations over time. The pairwise two sample t-test (including paired sample t-test and independent-samples t-test, P < 0.05) was conducted as post multiple comparison to reveal the changing trends of the two groups at the two time points. The relationship between differential connectivity and flight training hours in cadets was tested using Pearson’s correlation coefficient at a threshold of P < 0.05.
RESULTS
None of the subjects were excluded due to excessive head motion. Detailed information regarding the subjects is summarized in Table I. This study included 25 cadets and 24 healthy controls. There were no significant differences between the two groups in terms of age, handedness, or educational level.
The results of the mixed-model analysis are summarized in Table II (the intergroup and intragroup degrees of freedom are both 1, and the total degree of freedom is 48). There was a significant interaction in intermodular communication between parietal module and occipital module (FWE correction, P < 0.05). A pairwise, two-sample t-test was performed (including a paired sample t-test and an independent sample t-test, P < 0.05) (Table III) (the intragroup degree of freedom of cadet group was 24, the intragroup degree of freedom of control group was 23, and the intergroup degree of freedom is 47). In Table III, the first column represented the specific functional connections within or between modules, the remaining four columns represented the T-values of the functional connection differences in the two groups at two time points. Post hoc tests (including paired sample t-test and independent-samples t-test, P < 0.05) found that the connection between the parietal and occipital modules declined in the control group but improved greatly in cadets over the 2-yr period (Table III). In addition, there was no significant difference between the groups at the beginning because the group main effect did not exhibit any statistical significance. Over time, many alterations occurred in both groups (FWE correction, P < 0.05). Functional connectivity within the frontal module increased over the 2-yr period, while functional connectivity within the prefrontal module decreased in both groups. In addition, the interconnections between the frontal module and the prefrontal, parietal, temporal, and occipital modules increased in both groups. Meanwhile, the interconnection between the prefrontal and occipital modules and the interconnection between the subcortical and temporal modules decreased in both groups over the 2-yr period.
Correlation analysis examining the functional value of the brain and flight hours among cadets was performed. However, no significant correlations were found.
DISCUSSION
The present study investigated the effectiveness of flight training on a large-scale brain network. The interconnections between the parietal and occipital modules increased significantly in the cadet group. This alteration may be due to training effects rather than simply time. Although there were some alterations in both groups over time, these changes may be due to natural maturation or university studies.
Cadets at the CAFUC complete their theoretical studies in the classroom in the first 2 yr and complete their flight training within the final few years. Undergraduates of other majors only need to complete theoretical studies in the classroom. What these two groups have in common is that they learn theoretical knowledge in the classroom. This activity involves reading, memory, and executive function. In the current study, we identified that the functional connectivity within the frontal module increased over the 2-yr period, while functional connectivity within the prefrontal module decreased. The interconnections between the frontal module and the prefrontal, parietal, temporal, and occipital modules increased. Meanwhile, the interconnection between the prefrontal and occipital modules and the interconnection between the subcortical and temporal modules decreased over the 2 yr. All these changes were identified in both groups during the 2 yr. Thus, these alterations might represent the influence of time and study.
There was significant interaction in intermodular communication between the parietal module and occipital module. However, the data from the first assessment did not differ between the two groups. Therefore, different trends in connection between the parietal and occipital modules in the two groups may have occurred because the cadets learned flight operations whereas the controls did not.
The parietal module includes the paracentral lobule, postcentral gyrus, superior parietal gyrus, inferior parietal gyrus, angular gyrus, supramarginal gyrus, precuneus, and posterior cingulate gyrus. The occipital module includes the superior occipital gyrus, middle occipital gyrus, inferior occipital gyrus, cuneus, lingual gyrus, calcarine fissure, and surrounding cortex. The parietal region is known as the site of sensorimotor integration. This region can be divided into two major regions. The somatosensory cortex is located at the front. The paracentral lobule is the junction between the precentral and postcentral gyri on the surface of the medial hemisphere. It connects the precentral and postcentral gyri. Its function is related to motor and sensory control of the contralateral lower extremity.29 The postcentral gyrus is involved in somatosensory function and somatotopic representation of the body surface.27 The posterior parietal cortex is located at the junction of multiple sensory regions such as the visual, auditory, and tactile cortices. It has direct and indirect connectivity with the cortical and subcortical regions associated with motor responses. Due to its unique location, the parietal cortex appears to play a crucial role in transforming sensory inputs into motor outputs.4 Functional MRI data from humans and physiological studies involving monkeys suggest functional specialization in the parietal lobe in mapping between different sensory stimuli and different motor actions.1,28 Some regions of the intraparietal sulcus appear to be involved in the detection of new salient information and maintenance of attention over time.23 When there is a change in visual or auditory stimuli, the activation of the temporoparietal junction, located in the parietal lobe, is significantly enhanced, especially when the stimulus is relevant to the current behavior.15,16 The accuracy of movement depends on real-time sensory feedback. Evidence has shown that the sensory control of actions depends on the superior and inferior parietal lobules.11 Lesions in the inferior or superior parietal lobules lead to inaccurate reaching of the visual target.30 Studies using transcranial magnetic stimulation have found that disruption of posterior parietal cortex function led to inaccurate location and inability to learn new movement trajectories.13,14,34 Navigation relies on the coordinated activity of several distinct brain areas.
Evidence suggests that the parietal lobe is involved in visually guided grasping and visual–tactile integration.7,18,20 The parietal lobe can synthesize a variety of senses (especially visual information) to form spatial information for monitoring and adjusting movement. In conclusion, the parietal cortex is an association area for integrating different stimuli from different sensory channels. Its role in space perception and guiding actions has been emphasized.19 The increased functional connectivity between the parietal and occipital modules suggests improved hand–eye coordination in flight cadets.
In addition, in a virtual environment, the posterior parietal cortex was activated during active navigation task.32 Moreover, the medial parietal cortex was specially activated during learning heading.3 The angular and supramarginal gyrus, which belong to inferior parietal lobule, are known as key cortical areas that integrate vestibular inputs with other sensory information.2,24 The parieto-insular vestibular cortex, which involves parietal cortex, has been described to be the core region of the vestibular cortical system.21,22 The region around the intraparietal sulus always activates during eye movements.26 Caloric vestibular stimulation could cause activation of the parietal and occipital areas in humans.17 It seems that the parietal and occipital cortex are involved in vestibular signal processing. Spatial orientation and resistance to space motion sickness are critical in aviation. They are also an important part of flight training. In flight, pilots need to know exactly where they are. They are trained to judge their current position based on their feelings and instrument information. In addition, the cadets have to train on different kinds of looping (Jedrys test).35 The increased functional connectivity between the parietal and occipital modules might also reflect these effects.
Moreover, a subset of visuomotor neurons (mirror neurons) is activated during the observation or description of other individual actions. These mirror responses have been reported in the human parietal and frontal cortexes.31 The mirror system may play a crucial role in learning new actions.25 Studies have suggested that the parietal cortex contributes to action observation when the observer intends to imitate them later.5
Flight operations involve the real-time observation of instrument information inside the cockpit and environmental information outside the aircraft. The pilot needs to control the aircraft (attitude, heading, etc.) using the stick and pedal according to the current state of the aircraft displayed by the instrument panel; as such, flying requires a large amount of hand–eye coordination. Cadets also learn to fly by observing flight instructors’ operations. Such day-to-day learning may intensify the connection between the parietal and occipital modules among cadets.
The main limitation of this study was its small sample size. Owing to the limitations of the equipment and the difficulty of a longitudinal study, only 25 cadets and 24 controls were included. We did not find any relationship between the brain functional properties and flight training data, which may be due to the small sample size.
In conclusion, we investigated the effect of flight training on the functional brain networks of 25 cadets and 24 controls. The interconnections between the parietal and occipital modules increased significantly in the cadet group and decreased significantly in the control group. This alteration was likely due to >200 h of flight training performed by cadets.
The distribution of the six modules. Color represents the number of modules. 1) Frontal module; 2) prefrontal module; 3) subcortical module; 4) parietal module; 5) temporal module; and 6) occipital module.
Contributor Notes
