The Influence Of Personality Type On Depressive Symptoms And Its Neural Basis

The heterogeneity problem in mental health has always attracted much attention which restricted the research focus on mental health and cognitive neuroscience pointed out by some researchers.At present,a large number of empirical studies provide evidence that isolated personality traits are often associated with depressive symptoms.However,only a few studies consider the probable effect from a perspective of taxonomy.It has been proved that using a person-centered typological approach to investigate the relationship between personality and mental health problems is valuable due to the various types of personality associate differently with mental health.Previous studies have identified three personality types(RUO personality types):resilient,undercontrolled,and overcontrolled.Overcontroller displayed the most serious depressive symptoms.However,RUO classification has been widely challenged in statistics among the past two decades.First,the results obtained using different methods and samples could not identify over three types or could not be reproduced completely.Second,the studies recruited RUO classification didn‘t show much consistency.These arguments demonstrate that,to a certain extent,the RUO personality types lack consensus and replicability.Most of the current researches on personality types tend to adopt unsupervised clustering methods to identify subtypes.The unsupervised clustering method is ideal when used to explores possible personality types formed by all personality traits.When exploring personality types combined with specific questions or outcomes(such as combining the severity of depressive symptoms,or whether or not someone is diagnosed as depression),unsupervised clustering may not meet the needs.As different cognitive functions or clinical diseases might relate to different personality traits,the personality types combined with specific questions can correspond to various input combinations from different personality traits.The previous study has paid attention to the neural basis of unidimensional personality traits and their relation to the depressive symptoms instead of one multidimensional scale.Only a handful research explored the neural basis of personality types,even so,they almost missed out the potential relation underlying personality types and depressive symptoms.Moreover,understanding the relation of the two and the specific personality-depression theoretical model it embedded in has been rather limited.To solve the aforementioned issues,this thesis combines multi-modal brain imaging data,two longitudinal research designs,and multiple data analysis methods to conduct a systematic study on the relationship between personality types and depressive symptoms.First,we identified personality types associated with depression in non-clinical and clinical samples.In study 1,the NEO personality inventory and the Beck Depression Inventory of1136 college students(sample 1)were collected.Then,combined with the severity of depression symptoms and adopted the conditional inference tree(supervised machine learning classification method),we identified personality types associated with depression in a healthy population: two stable types,three vulnerable types,and three resilient types.The classification result was validated on another sample containing 265 college students(sample 2),and the predictive power of personality types and personality traits on depressive symptoms was compared.Our results indicated that the predictive power from personality types was better than that of unidimensional personality traits.In study 2,personality data of 135 depression patients and 133 controls(sample 3)were collected.Combined with "depression diagnosis",the personality types of depressive patients were identified through the random forest and community detection(a combination of supervised and unsupervised machine learning classification method).It turned out that there were three vulnerable types and one resilient type in patients.This part of the research,combined the classification process of personality types with the severity of depression symptoms and the diagnosis of depression,identified the personality types which associated with depression in non-clinical samples and clinical samples respectively.Moreover,typical personality types showed similar tendency were also found in these two samples: vulnerable type of high neuroticism with low extraversion level(high depression risk),and resilient type of low neuroticism with high extraversion level(low depression risk).Next,we explored the differences in brain structure and function between the vulnerable type and the resilient type.In study 3,we collected 3D-T1 structural imaging and resting-state functional imaging data from sample 1 and sample 3.After quality control,the structural imaging data of 932 participants and the resting functional image data of 1071 participants were available in sample 1;the structural image and resting-state image of 131 participants and 128 controls were available in sample 3.Then we selected 14 subcortical areas as regions of interest and compared the difference of subcortical volume between the vulnerable type and the resilient type in the two samples.The results showed that there was no difference in the subcortical volume between the two,but the volume of the right nucleus accumbens of the patient vulnerable type was smaller than that of the control vulnerable type.Then we selected nodes of the subcortical network as regions of interest according to the power-264 template and calculated the functional connectivity map of the region of interest to the whole brain.Based on the functional connectivity map,we compared the differences of resting-state functional connectivity between the resilient type and the vulnerable type.In sample 3,we further compared the difference on the resting-state functional connectivity of patient resilient type with control resilient type,and patient vulnerable type with control vulnerable type.Our analysis revealed that,in the healthy population,the functional connectivity strength between the left/right putamen and the right superior parietal gyrus,and between the right thalamus and several nodes in the visual/somatomotor network were significantly weaker in the vulnerable type than in the resilient type.In depression patients,the functional connectivity strength between the putamen/ thalamus and the default network was significantly stronger in the vulnerable type than in the resilient type.The patient resilient type was weaker than the control resilient type on the functional connectivity strength which largely distributed between the subcortical area and the default network/frontoparietal network/limbic system,and a small amount was distributed between the subcortical area and the ventral attention network/somatomotor network/visual network.In study 4,we recruited 22 vulnerable participants and 35 resilient participants in sample 1 and collected the emotional regulation task-f MRI data(cognitive reappraisal paradigm).Then,we explored the different cognitive neural basis in emotional processing between the vulnerable type and the resilient type.We further compared the activity differences of different task conditions and compared the functional connectivity differences under different conditions based on the physiological and psychological interaction(PPI)analysis(brain nodes in the subcortical network as regions of interest were also used in the process).The results found that the vulnerable type showed weaker activation in the postcentral gyrus than the resilient type under the negative affect state(viewing negative affect pictures > viewing neutral affect pictures).PPI analysis showed that the vulnerable type displayed weaker functional connectivity between the thalamus and the postcentral gyrus,and between the putamen and the precentral gyrus than the resilient type,while the vulnerable type demonstrated stronger functional connectivity between the midbrain and the inferior parietal lobule.Besides,under the state of cognitive reappraisal processing(decrease negative affect > viewing negative affect pictures),the vulnerable type performed stronger functional connectivity between the thalamus and the middle frontal gyrus and between the putamen and the middle temporal gyrus,whereas the vulnerable type showed weaker functional connectivity between the putamen and precentral gyrus than the resilient type.Finally,we conducted continuous longitudinal tracking and used three studies to further investigate the influence of personality types on depressive symptoms and neural basis.In study 5,we collected 138 participants‘ personality and depressive symptoms data at two time points,and by adopting the latent change score model we aimed to resolve three issues: 1.Whether initial depressive symptoms can predict the change of individual personality type;2.Whether the initial personality type can predict the change of individual depressive symptoms;3.Whether the change of depressive symptoms is related to the change of personality type.The results showed that initial personality type affected the change of depressive symptoms,but initial depressive symptoms didn‘t affect the change of personality type.There was a co-variant relationship between the change of personality type and the change of depressive symptoms.The results of study 5 showed that personality type can affect the change of depressive symptoms,but it is not clear whether personality type can affect the rate of change in depressive symptoms.Therefore,study 6 sought to address this question.By tracking 99 participants at three different times,we obtained the personality data of the participants at time point 1 and the depressive symptoms data at time points 1,2,and 3.Then,we tried to fit the developmental trajectory of the depressive symptoms,the latent growth curve model and the latent growth mixture model were used in the fitting process.Finally,in the framework of structural equation,the initial personality type was used to predict the developmental trajectory of depressive symptoms.The results showed that the initial personality type can well predict the developmental trajectory(rate of change)of depressive symptoms.In study 7,we performed structural and functional imaging scans of the participants who completed study 6 and obtained the personality and brain structural and functional data of the participants at two time points.Through the latent change score model,three questions were elucidated: 1.Whether the initial personality type can predict the change of brain structure and function;2.Whether the initial brain structure and function can predict the change of personality type;3.Whether the change of personality type is related to the change in brain structure and function.The results showed that the initial personality type predicted the change of the functional connectivity between the right putamen and the superior parietal gyrus,but the initial brain structure and function cannot predict the change of the individual personality type,and there was no co-variant relationship between the change of personality type and the change of brain structure and function.In summary,the personality types associated with depression are mainly composed of a combination of neuroticism and extraversion.There are two typical personality types in both healthy population and depressive patients: the vulnerable type with high depression risk(a combination of high neuroticism and low extraversion)and the resilient type with low depression risk(a combination of low neuroticism and high extraversion).The neuroticism and extraversion level of other personality types are mainly between the continuum of the vulnerable and the resilient type.In terms of brain structure,the volume of the right nucleus accumbens of the patient vulnerable type is smaller than that of the control vulnerable type.In terms of brain function,compared to the resilient type,the strength of the functional connectivities between the putamen and superior parietal gyrus,the thalamus and the postcentral gyrus,the putamen and the precentral gyrus,as well as the thalamus and several nodes in the visual/somatomotor network is significantly weaker in the vulnerable type.The functions of these functional connectivities might be related to cognitive control,attentional control and executive control.This may indicate that the vulnerable type has a weaker ability of top-down constraint.Comparing to the resilient type,the strength of the functional connectivities between the midbrain and the inferior parietal lobule,the thalamus and the middle frontal gyrus,the putamen and the middle temporal gyrus,as well as the putamen/thalamus and several nodes in the default network is significantly stronger in the vulnerable type.The functions of these functional connectivities might be related to self-related processing and self-focus,especially facing negative stimuli.This may indicate that the vulnerable type has a stronger bias in the processing of self-related information.Although the same resilient type,the patient resilient type still exhibits a weaker top-down constraint involving most of the cortical networks.Importantly,personality type can predict the degree of change and developmental trajectory of depressive symptoms,and can also predict the change of functional connectivity which is related to top-down constraint.Our results more support the precursor model or predisposition model.The thesis associated personality type with the severity of depressive symptoms and the diagnosis of depression,and abstracts individuals into an effective classification system.Our findings can contribute to identifying specific depression populations on different degrees(especially high-risk populations)to facilitate early intervention.Additionally,it helps to reveal the neural basis of different personality types and depression.More importantly,we provided researching basis on improving mental health status.