How Functional Connectivity Measures Affect the Outcomes of Global Neuronal Network Characteristics in Patients with Schizophrenia Compared to Healthy Controls.

Modern computational solutions used in the reconstruction of the global neuronal network arrangement seem to be particularly valuable for research on neuronal disconnection in schizophrenia. However, the vast number of algorithms used in these analyses may be an uncontrolled source of result inconsistency. Our study aimed to verify to what extent the characteristics of the global network organization in schizophrenia depend on the inclusion of a given type of functional connectivity measure. Resting-state EEG recordings from schizophrenia patients and healthy controls were collected. Based on these data, two identical procedures of graph-theory-based network arrangements were computed twice using two different functional connectivity measures (phase lag index, PLI, and phase locking value, PLV). Two series of between-group comparisons regarding global network parameters calculated on the basis of PLI or PLV gave contradictory results. In many cases, the values of a given network index based on PLI were higher in the patients, and the results based on PLV were lower in the patients than in the controls. Additionally, selected network measures were significantly different within the patient group when calculated from PLI or PLV. Our analysis shows that the selection of FC measures significantly affects the parameters of graph-theory-based neuronal network organization and might be an important source of disagreement in network studies on schizophrenia.

Automated Classification of Cognitive Workload Levels Based on Psychophysiological and Behavioural Variables of Ex-Gaussian Distributional Features.

The study is focused on applying ex-Gaussian parameters of eye-tracking and cognitive measures in the classification process of cognitive workload level. A computerised version of the digit symbol substitution test has been developed in order to perform the case study. The dataset applied in the study is a collection of variables related to eye-tracking: saccades, fixations and blinks, as well as test-related variables including response time and correct response number. The application of ex-Gaussian modelling to all collected data was beneficial in the context of detection of dissimilarity in groups. An independent classification approach has been applied in the study. Several classical classification methods have been invoked in the process. The overall classification accuracy reached almost 96%. Furthermore, the interpretable machine learning model based on logistic regression was adapted in order to calculate the ranking of the most valuable features, which allowed us to examine their importance.

Recognition of Electroencephalography-Related Features of Neuronal Network Organization in Patients With Schizophrenia Using the Generalized Choquet Integrals.

In this study, we focused on the verification of suitable aggregation operators enabling accurate differentiation of selected neurophysiological features extracted from resting-state electroencephalographic recordings of patients who were diagnosed with schizophrenia (SZ) or healthy controls (HC). We built the Choquet integral-based operators using traditional classification results as an input to the procedure of establishing the fuzzy measure densities. The dataset applied in the study was a collection of variables characterizing the organization of the neural networks computed using the minimum spanning tree (MST) algorithms obtained from signal-spaced functional connectivity indicators and calculated separately for predefined frequency bands using classical linear Granger causality (GC) measure. In the series of numerical experiments, we reported the results of classification obtained using numerous generalizations of the Choquet integral and other aggregation functions, which were tested to find the most appropriate ones. The obtained results demonstrate that the classification accuracy can be increased by 1.81% using the extended versions of the Choquet integral called in the literature, namely, generalized Choquet integral or pre-aggregation operators.

On the Improvement of Eye Tracking-Based Cognitive Workload Estimation Using Aggregation Functions.

Cognitive workload, being a quantitative measure of mental effort, draws significant interest of researchers, as it allows to monitor the state of mental fatigue. Estimation of cognitive workload becomes especially important for job positions requiring outstanding engagement and responsibility, e.g., air-traffic dispatchers, pilots, car or train drivers. Cognitive workload estimation finds its applications also in the field of education material preparation. It allows to monitor the difficulty degree for specific tasks enabling to adjust the level of education materials to typical abilities of students. In this study, we present the results of research conducted with the goal of examining the influence of various fuzzy or non-fuzzy aggregation functions upon the quality of cognitive workload estimation. Various classic machine learning models were successfully applied to the problem. The results of extensive in-depth experiments with over 2000 aggregation operators shows the applicability of the approach based on the aggregation functions. Moreover, the approach based on aggregation process allows for further improvement of classification results. A wide range of aggregation functions is considered and the results suggest that the combination of classical machine learning models and aggregation methods allows to achieve high quality of cognitive workload level recognition preserving low computational cost.

Manipulations of the Response-Stimulus Intervals as a Factor Inducing Controlled Amount of Reaction Time Intra-Individual Variability.

Aggrandized fluctuations in the series of reaction times (RTs) are a very sensitive marker of neurocognitive disorders present in neuropsychiatric populations, pathological ageing and in patients with acquired brain injury. Even though it was documented that processing inconsistency founds a background of higher-order cognitive functions disturbances, there is a vast heterogeneity regarding types of task used to compute RT-related variability, which impedes determining the relationship between elementary and more complex cognitive processes. Considering the above, our goal was to develop a relatively new assessment method based on a simple reaction time paradigm, conducive to eliciting a controlled range of intra-individual variability. It was hypothesized that performance variability might be induced by manipulation of response-stimulus interval's length and regularity. In order to verify this hypothesis, a group of 107 healthy students was tested using a series of digitalized tasks and their results were analyzed using parametric and ex-Gaussian statistics of RTs distributional markers. In general, these analyses proved that intra-individual variability might be evoked by a given type of response-stimulus interval manipulation even when it is applied to the simple reaction time task. Collected outcomes were discussed with reference to neuroscientific concepts of attentional resources and functional neural networks.

Interpretable Machine Learning Models for Three-Way Classification of Cognitive Workload Levels for Eye-Tracking Features.

The paper is focussed on the assessment of cognitive workload level using selected machine learning models. In the study, eye-tracking data were gathered from 29 healthy volunteers during examination with three versions of the computerised version of the digit symbol substitution test (DSST). Understanding cognitive workload is of great importance in analysing human mental fatigue and the performance of intellectual tasks. It is also essential in the context of explanation of the brain cognitive process. Eight three-class classification machine learning models were constructed and analysed. Furthermore, the technique of interpretable machine learning model was applied to obtain the measures of feature importance and its contribution to the brain cognitive functions. The measures allowed improving the quality of classification, simultaneously lowering the number of applied features to six or eight, depending on the model. Moreover, the applied method of explainable machine learning provided valuable insights into understanding the process accompanying various levels of cognitive workload. The main classification performance metrics, such as F1, recall, precision, accuracy, and the area under the Receiver operating characteristic curve (ROC AUC) were used in order to assess the quality of classification quantitatively. The best result obtained on the complete feature set was as high as 0.95 (F1); however, feature importance interpretation allowed increasing the result up to 0.97 with only seven of 20 features applied.

Resting-state hyperconnectivity within the default mode network impedes the ability to initiate cognitive performance in first-episode schizophrenia patients.

Among multiple cognitive impairments present in schizophrenia, a decline in fast information processing is one of the most severe neuropsychological deficit. Reduced ability to efficiently launch a coherent cognitive activity might be a significant factor contributing to poor results in time-limited tasks obtained by schizophrenia patients. The aim of this study was to identify neurophysiological predictors of expected cognitive initiation failures in a group of first-episode schizophrenia individuals (SZ). To evaluate the effectiveness of initiation, a dynamic analysis of design fluency test was applied, assessing to what extent the productivity was focused within the first interval of the performance, what is a typical way healthy subjects execute this task. Resting-state EEG recordings were obtained from SZ patients (n = 34) and controls (n = 30) to examine functional connectivity between 84 intra-cortical current sources determined by eLORETA (exact low-resolution brain electromagnetic tomography) for six conventionally analyzed frequencies. The nonparametric randomization approach was used to identify hypo- and hyper-connections, i.e. synchronizations significantly differentiating the studied samples in terms of connectivity strength. Generally, SZ patients obtained poor outcomes in fluency test and dynamic analysis of performance confirmed the presence of initiation deficit in clinical sample, which was a single factor explaining the intergroup difference regarding the entire task. In the majority of frequencies, the arrangement of synchronizations in SZ group was dominated by hypo-connections, except for the theta band, in which the strength of synchronizations between posterior cingulate cortex, cuneus and precuneus was significantly higher for SZ group. These theta-band hyper-connections turned out to be significant predictors of cognitive initiation failure in the clinical sample. Additionally, theta hyper-connections correlated negatively with the total number of unique designs generated by patients, however, the strength of this correlation was weaker than regarding initiation index. The results of this study suggest that baseline hyperconnectivity within the posterior hub of the Default Mode Network, containing posterior cingulate gyrus and precuneus, might disturb effective cognitive outcome, not only by interfering with task-positive functional networks but also by delaying the starting phase of performance, which might be specifically deleterious for the execution of time-limited tests.

Cognitive inconsistency in bipolar patients is determined by increased intra-individual variability in initial phase of task performance.

BACKGROUND: Bipolar patients show high intra-individual variability during cognitive processing. However, it is not known whether there are a specific fluctuations of variability contributing to the overall high cognitive inconsistency. The objective was to compare dynamic profiles of patients and healthy controls to identify hypothetical differences and their associations with overall variability and processing speed. METHODS: Changes of reaction times iSD during processing speed test performance over time was measured by dividing the iSD for whole task into four consecutive parts. Motor speed and cognitive effort were controlled. RESULTS: Patients with BD exhibited significantly lower results regarding processing speed and higher intra-individual variability comparing with HC. The profile of intra-individual variability changes over time of performance was significantly different in BD versus HC groups: F(3, 207)=8.60, p<0.0001, ηp2=0.11. iSD of BD patients in the initial phase of performance was three times higher than in the last. There was no significant differences between four intervals in HC group. Inter-group difference in the initial part of the profiles was significant also after controlling for several cognitive and clinical variables. LIMITATIONS: Applied computer version of Cognitive Speed Test was relatively new and, thus, replication studies are needed. Effect seen in the present study is driven mainly by the BD type I. CONCLUSIONS: Patients with BD exhibits problems with setting a stimulus-response association in starting phase of cognitive processing. This deficit may negatively interfere with the other cognitive functions, decreasing level of psychosocial functioning, therefore should be explored in future studies.