Greeshma Sharma

Eye tracking based human computer interaction: Applications and their uses

With the evolution of Eye Tracking from a concept to reality, it is being explored scientifically these days in Human Computer Interaction in order to record the eye movements to determine the gaze direction, position of a user on the screen at a given time and the sequence of their movement. The threefold objective of this paper include introducing the reader to the key aspects and issues of eye-movement technology, practical guidance for developing an Eye tracking application, and various opportunities and underlying challenges to develop (Man and Machine Interfacing) MAMI systems using Eye tracking. We have uniquely integrated The Eye Tribe with Unity5.1.1 and through an experiment, we have also inferred that a subject with and without bifocal glasses show relatively similar fixation results if they have correct vision but the results differ with small error if the eye is corrected using lenses. Another experiment using Eye Tribe shows that gaze input requires less time as compared to the mouse input.

Effect of Sudarshan Kriya (meditation) on gamma, alpha, and theta rhythm during working memory task.

Aims: The present study focuses on analyzing the effects of Sudarshan Kriya yoga (SKY) on brain signals during a working memory (WM) task. To envision the significant effects of SKY on WM capacity (WMC), we chose a control group for contriving a cogent comparison that could be corroborated using statistical tests. Subjects and Methods: A total of 25 subjects were taken in the study, of which 10 were allotted to a control group and 15 to an experimental group. Electroencephalograph was taken during a WM task, which was an automated operation span test before and after SKY with 90 days intervals. No SKY was given to the control group. Statistical Analysis Used: t-test and one-way ANOVA were applied. Results: SKY promoted the efficient use of energy and power spectral density (PSD) for different brain rhythms in the desired locations as depicted by the gamma (F8 channel), alpha, and theta 2 (F7 and FC5) bands. It was found that gamma PSD reduced for both phases of memory in the experimental group. Alpha energy increased during the retrieval phase in the experimental group after SKY. Theta 1 rhythm was not affected by SKY, but theta 2 had shown left hemispheric activation. Theta rhythm was associated with memory consolidation. Conclusions: SKY had shown minimized energy losses while performing the task. SKY can improve WMC by changing the brain rhythms such that energy is utilized efficiently in performing the task.

Influence of Landmarks on Wayfinding and Brain Connectivity in Immersive Virtual Reality Environment

Spatial navigation is influenced by landmarks, which are prominent visual features in the environment. Although previous research has focused on finding advantages of landmarks on wayfinding via experimentation; however, less attention has been given to identifying the key attributes of landmarks that facilitate wayfinding, including the study of neural correlates (involving electroencephalogram, EEG analyses). In this paper, we combine behavioral measures, virtual environment, and EEG signal-processing to provide a holistic investigation about the influence of landmarks on performance during navigation in a maze-like environment. In an experiment, participants were randomly divided into two conditions, Landmark-enriched (LM+; N = 17) and Landmark-devoid (LM-; N = 18), and asked to navigate from an initial location to a goal location in a maze. In the LM+ condition, there were landmarks placed at certain locations, which participants could use for wayfinding in the maze. However, in the LM- condition, such landmarks were not present. Beyond behavioral analyses of data, analyses were carried out of the EEG data collected using a 64-channel device. Results revealed that participants took less time and committed fewer errors in navigating the maze in the LM+ condition compared to the LM- condition. EEG analyses of the data revealed that the left-hemispheric activation was more prominent in the LM+ condition compared to the LM- condition. The event-related desynchronization/synchronization (ERD/ERS) of the theta frequency band, revealed activation in the left posterior inferior and superior regions in the LM+ condition compared to the LM- condition, suggesting an occurrence of an object-location binding in the LM+ condition along with spatial transformation between representations. Moreover, directed transfer function method, which measures information flow between two regions, showed a higher number of active channels in the LM- condition compared to the LM+ condition, exhibiting additional wiring cost associated with the cognitive demands when no landmark was available. These findings reveal pivotal role of the left-hemispheric region (especially, parietal cortex), which indicates the integration of available sensory cues and current memory requirements to encode contextual information of landmarks. Overall, this research helps to understand the role of brain regions and processes that are utilized when people use landmarks in navigating maze-like environments.

Sex differences in mental rotation: Cortical functional connectivity using direct transfer function

Abstract Previous studies have reported that sex differences exist in mental rotation (MR) through different activated cortical regions, but it remains unclear what could be possible reasons of such differences in the different processing stages of MR. A few Event related potential (ERP) studies have noticed that sex differences occur in relatively early cognitive processing stages, but none of the study has viewed directional flow of information in the earlier stages as a function of complexity in men and women. This study investigated possible reasons for sex differences in visuospatial performance by flux of information underlying cortical functional connectivity. In the present study, earlier two stages were identified as a) perceptual encoding, identification, and discrimination of objects, kept under visuospatial attention allocation network (VSAN) and b) rotation ability involving spatial transformation strategy, assigned in mental rotation network (MRN). Participants underwent 3D mental rotation task with varying difficulty levels, simultaneously having electroencephalogram (EEG). It has been confirmed in behavioural outcome, as angular disparity increases, reaction and accuracy trades off. There were different activated electrodes in male and female participants for both networks. Advantage of spatial working memory was evident in men and reflected during performance. Also, VSAN showed that men utilised bottom-up attentional processes for more rotated views. MRN exhibited hemispheric lateralisation in the parietal cortex; men showed higher activation in right parietal cortex. This research work offers promising perspective to the study of cortical functional connectivity, in the terms of strength and direction, during sub-processes of MR.

Corrigendum: Influence of Landmarks on Wayfinding and Brain Connectivity in Immersive Virtual Reality Environment

[This corrects the article on p. 1220 in vol. 8, PMID: 28775698.].

Playing Action Video Games a Key to Cognitive Enhancement

Abstract In this paper, we aim to analyse the impact of training on improvement in cognitive abilities and performance of the subjects playing single player action video game. Recent research indicates that playing Action video games improve cognitive abilities. However no study has exploited the novel technique, Empirical Mode Decomposition in the field of action video games. Empirical mode decomposition was used to extract various features by decomposing EEG data into intrinsic mode functions. Intrinsic mode functions were used to calculate linear features like standard deviation, phase and energy. K- Nearest Neighbour & Linear Discriminant Analysis classifiers were used to classify the subject based on the changes in features extracted due to the impact of training. Psychological tests conducted before and after the training, positively affirm that training improves cognitive abilities like reaction time and reduces stress level.

Capacitating surveillance and situational awareness with measure of visual engagement using eyetracker

Military application requires continuous surveillance and situation awareness at high security zones. For maintaining such requirements, there is a need to monitor visual attention for a defense personnel. Eye movements, captured through eye tracker can reveal visual engagements with the help of examining the direction of eye gaze. This not only reveals the opinion and behavioral patterns but also gives an idea that the brain can reveal hidden and profitable truths about consumers. Henceforth making it applicable in defense as well as in neuromarketing, as both domain want to grab visual attention of their personnel. This study tries to assess visual engagements through eye tracker using neuromarketing paradigm. Results can further be extended to the defense areas both fields involve visual attention. This paper investigates the use of Eyetracking as a potential research tool for analyzing the features that drives attention towards certain target (advertisement in this case). It is believed that eye movements are best indicators of visual attention. This paper gives an insight on the features that help us analyze visual attention and retention as tools to distinguish the effectiveness of the advertisement. The best feature of eye movements captured from eye tracking can identify glimpse of visual attention. Thus, making it greatly important in the use of defence as visual attention and henceforth brain reflexes are effected at immediate situations making it useful for surveillance, required feature in situational awareness and surveillance for border. This work here is of potential use in defence as it gives the measures of opinion and behavior of brain patterns of subjects revealing their hidden and profitable truths as that of customers.

Non-linear EEG correlates of route learning in virtual MAZE

To quantify route learning with the help of electroencephalogram (EEG) parameters during navigation is challenged with the individual differences. The correlation of non linear EEG features is identified and Hjorth parameter and Shannon Entropy computed for gamma and theta frequency bands, with the route learning in the virtual maze. EEG is a non-stationary signal which can be best recognised by non linear features and can detect precise change in EEG dynamics for learning perspectives. Subsequently good and bad performers are classified using behavioural scoring and its cross validation is performed by SVM classifier for the selected non-linear EEG features. Virtual maze task consists of two stages, i.e. stage1 and stage2, having differences only in the maximum allotted time for navigation. It is revealed from the results that Hjorth Parameter is more sensitive to change in learning as compared to Shannon Entropy. However, both the features are increased in stage 2, as the indicator of performance is improved. Furthermore, classification between good and bad performer show 62.5% accuracy for all extracted features while 80% accuracy is obtained in case of Hjorth parameter. The result suggests that Hjorth parameter is most correlated feature with route learning and can be used for discriminating individual on the basis of performance.

Influence of Spatial Learning Perspectives on Navigation Through Virtual Reality Environment

In navigation with virtual reality, spatial knowledge can be acquired through both route and survey perspective. Our study correlates the influence on spatial knowledge while navigating in a virtual reality environment after gaining information with different spatial perspectives. We measured brain activations while the participants navigated through a complex spatial environment, using the analysis tool of sLORETA. In the experimental condition, the participant watched a simulated video feed of either route perspective (front view) or survey perspective (top view) of the virtual environment. Distance travelled, path efficiency and time efficiency of the participants were measured while they navigated through nine successive landmarks. We obtained significant differences between the brain activation patterns while comparing both conditions. Higher activations in inferior frontal gyrus, parahippocampal gyrus, superior temporal gyrus and insula were observed for the theta band in route perspective when compared to survey perspective. Higher activations in the inferior parietal lobule, angular gyrus and precuneus were observed in survey perspective when compared to route perspective. Results showed higher path efficiency and time efficiency and lower distance travelled to reach the destination in survey perspective when compared to route perspective. The result indicates that survey perspective is better for navigation in a far spaced virtual reality environment.

Brain connectivity in spatial orientation task

Spatial orientation (SOT) is one of the spatial skill which is required in motor imagery and navigation. A lot of researches have shown underlying neural dynamics during SOT as an effect of spatial transformation strategy, but the consistent findings related to flow of information between designated brain areas are not present. The purpose of the study is to identify flux of information between pairs of channel to identify active hubs during the course of task. The directed transfer function (DTF) method is used to detect alterations in the functional coupling of EEG rhythms (0.5–30 Hz) in different brain cortical areas during virtual reality based perspective taking test. Results confirmed that there were common regions for motor imagery and SOT. It has also shed light on the active role of parietal and occipital lobe in SOT. This study proffers DTF as a useful technique for identifying hubs and flux of information in neuroscience.