Brain Dynamics in the Frontal Midline

The results showed the action of ICA applied to EEG data to remove effects of non-brain activities from scalp recordings. We classified the results of the ERSP based on path ratio categories (Figure 4-7). The power and amplitude of the theta band (4~8Hz) in the global condition were obvious strong. The theta power continuously increased in both global condition and local condition. When females searched the targets, the duration time of theta power increased longer in global condition than in local condition.

(a)Path ratio=1

(b) 1≤ Path ratio 1.5 ≤

(c) 1.5≤ Path ratio

Figure 4-7: (a) Path ratio=1 (b) 1≤ Path ratio ≤ 1.5 (c) 1.5 ≤ Path ratio

The theta power continuously increased in both global condition and local condition.

4.2.2. EEG Activites in Different Independent Components

4.2.2.1 Independent Component(IC) Clustering and Source Localization

The component clusters showed that the spatial navigation involved frontal midline components.

Figure 4-8: The frontal IC clusters of the location.

4.2.2.2. Frontal Midline Component

The block contained four states in the searching situation. The time range was at most twenty one seconds in a task, but the times of blocks were very variable.

Therefore we adjusted the time scale of the navigation time in order to compare brain activity. We enlarged the time less than ten seconds and narrowed down beyond ten seconds when subjects searched the target. The time of static state was fixed so that the time was adjusted sixteen seconds in a block. Figure 4-8 showed the ERSP results in the frontal midline. In the first state, subjects were jumped to the starting point and

the environment did not cause obvious brain activity response for subjects. In the second state, the target was appeared for subjects to search and the brain activity of theta power was increased in the last one second. We wanted subjects to recall where the target was and the phenomenon involved with working memory in the static state.

Obviously, the theta power lasted a long time in the third state when subject searched the target in the virtual reality environment. In the fourth state, the theta power returned like as the baseline power. The duration time and strength of theta power lasted longer in global condition than in local condition. Females recalled the location of the target in the early searching time so that the theta power increased and the brain activity of subjects became weak in the end of searching time. The peak value of the theta power was appeared in the third state in which females searched the target. The theta power was in the peak value for a long time in the global condition but the theta power was in the peak value for a short while in the local condition.

(a)

(b)

Figure 4-9: (a) Brain activity of theta power increased in frontal midline in the global condition. (b) Brain activity of theta power increased in frontal midline in the local condition.

4.2.2.3. Parietal Component

The parietal lobe integrated sensory information from different modalities, particularly determining spatial sense and navigation. We could observe that power decreased in the parietal during the experiment. In the first state, subjects exchanged the location in the beginning and power decreased for a while. The target was appeared in the second state and power decreased across almost second state. In the third state, females searched the target in the virtual environment and power decreased across the whole state. Figure 4-9 showed the results of ERSP in the parietal.

(a)

(b)

Figure 4-10: (a) Brain activity of power decreased in the parietal in the global condition. (b) Brain activity of power decreased in the parietal the in the local condition.

4.2.3. Brain Activity of Permutation Tests

Figure 4-10 showed the brain activity of power variation and significant differences between global condition and local condition. The theta band increased values during the experiment with different landmarks. There were significant differences of theta band between global condition and local condition. In addition, the theta power was higher in the global condition than in the local condition during the experiment.

Figure 4-11: The results of permutation tests indicated the significant differences in the gray block. Brain activity of theta power increased and had significant differences in the frontal midline.

.

4.3. Summary of Brain Activity Experiment

Females would be trained to reach 100% success rate before the brain dynamic experiment and females were familiar with the virtual environment. We observed the behavior performances in the brain dynamic experiment and females had better performance of path ratio in the global condition than in the local condition during the

experiment for each trial. The result of the performance was the same as behavioral experiment for females in comparison of path ratio. Global landmarks seemed more helpful for females to move in the virtual environment.

We classified the path ratio into three levels of path lengths which were short, medium, and long path. Females had more percentage of trials that were consisted of short and medium path length in the global condition than in the local condition. For this reason, females would not waste too much redundant path walking in the global condition.

The increase of the theta band was present in frontal midline. We observed that theta power increased during the navigation task. The theta band was more power and strength in global condition than in local condition. In permutation tests, there were significant differences of theta band between the global condition and the local condition during navigation in the frontal midline. In addition, we also observed that power decreased in the parietal during the experiment.

V. Discussion

5.1 Discussion of Behavior Results

The cognitive styles might be regarded as individual’s consistent approach to process the information during recall and thinking [50]-[51]. The relationships of styles were observed behaviors, such as learning performance, learning preferences, subject preferences [52]. Conventional training design methodologies played the important role so that we designed the training state and testing state for different purposes. Subjects had sufficient searching time when subjects searched the target in the training state. The accommodation of spatial navigation in the training design process had the potential to improve the efficiency of individual performance. The process could be seen the cognitive styles of navigation. We determined the performance of subjects in the testing state. The males needed fewer learning trials (2.77±0.39) than females (4.5±0.39) to fulfill 100% success rate. The phenomenon appeared that males were more familiar with the virtual reality environment than females in the early trials.

The cognitive functions for gender-specific performance difference were well known [53]-[54]. The behavioral research demonstrated that women rely predominantly on landmark cues, whereas men use both geometric and landmark cues [55]. For previous research, males were better spatial perception ability than females in the spatial experiment [56]. Both types of landmarks might be provided with different help of guide for subject and both types of landmarks differenced dimension that affected strategy choice. Subjects could use the landmarks as references or make choices with respect to moving in the environment. The global landmarks were used as direct goals, whereas the local landmarks were used as intermediate goals. We observed the improvement of the searching time and path the ratio in the behavioral

experiment.

We also found the different performances of gender. In comparison of searching time, males spent less searching time than females for each trial in both types of landmarks. Males had better performance of searching time than females. Subjects sometimes needed to rotate their view to decide the direction in order to search the target so that ability of rotation skills possessed influence about searching time. In comparison of path ratio with respect to gender, females had a little better performance of corrected travel distance than males in the local landmarks environment. The phenomenon was appeared that males wasted more redundant walking path than females in the local landmarks environment. Females effectively adopted the local landmarks to navigate the virtual maze. Males and females had similar corrected path length in the global condition. In global landmarks environment, all males and females effectively made use of global landmarks to determine the direction during navigating. The closest landmarks around the target played the important role for subjects. In global condition, subjects could easily know where they were for a short while in the maze but this was not easy matter for subjects to know the location in the local condition. When subjects decided the direction to walk in the global condition, subjects would quite easily choose the shortest path. Therefore, subjects would not waste redundant walking in the global condition.

The empirical navigation was depended on practice and the different strategies were affected searching style. We also determined the detailed performance for males or females in the different condition. Males spent the similar amount of searching times in both types of landmark environments. Females spent less amount of searching time in the local than in the global condition. In comparison of path ratio with respect to landmarks, both of males and females had shorter corrected travel

distance in the global condition than in the local environment. The phenomenon appeared that the global landmarks played the role as beacons to guide the subjects the direction. As a result, subjects would not waste too many redundant walking paths form the information of the global landmarks. To sum up the females’ performance was different between searching time and path ratio. Subjects would waste the searching time which was constructed from walking time, rotating time and recalling time. The searching time could not absolutely present the ability of navigation. We considered the improved trend of the searching time. Females got better and better performance of the searching time. The path ratio presented the familiarity with the virtual environment.

Because subjects were randomly transformed to another position far away from the target in the first state, then even the same target would not be necessarily corresponded to the same initial position. For this reason, subjects might have different walking path in the different navigating task and the performance of subjects was slight ups and downs. The overall trend gradually progressed.

After subjects completed the experiment, subjects were asked to write the questionnaire. We found that all subjects could not be able to describe the integrant structure of the landmarks’ location about both type landmarks. Subjects could mark a part of landmarks surrounding target in the global condition. Subjects could only describe the landmarks that were close to the target in the local condition. Subjects only described the relative position between the target and landmarks. It was difficult for subjects to describe the entire map with both landmarks. It seemed that subjects memorized the landmarks related to the target and subjects played less attention to other landmarks. There was different weight of landmarks according to targets. The landmarks closed to the target possessed the most significant help so that subjects

memorized these relative landmarks to search the target. The other landmarks played important role of assistance to cognize where subjects was.

In the EEG experiment, females had been familiar with the virtual environment.

Females had shorter corrected travel distance in the global condition than in the local environment, too. It was obviously that there was stable performance in the global condition. Females still wasted redundant walking path in the local condition.

5.2 Discussion of Brain Activity

Theta oscillation could hence be conceived as the navigation rhythm through both physical and mnemonic space [57]. The theta band associated with states of concentration and conscious control over attention [58]. The observation of human theta had begun to reveal an intriguing connection between brain oscillations and cognitive processes [59] Both the frontal cortex and the centroparietal cortical areas contributed to learning during spatial navigation [60]. Prominent theta activity appeared over the frontal midline. The theta rhythm in the frontal midline increased in magnitude with memory load and where as a parietal central, alpha signal decreased in magnitude with increased task difficulty [21]. Previous researches demonstrated that the theta power in the frontal midline scalp increased with mental effort associated with working memory [61]-[64]. Memory recall had been associated with an increase of theta band during encoding and retention [65]-[66]. Some studies [67]-[68] reported that theta of the frontal midline appeared during short-term working memory tasks and the theta activity increases with memory load. The theta band in the scalp in the frontal midline was often present during working memory [29]. In the EEG experiment, we observed the increase of the theta (4-8 Hz) power in the frontal midline. Before the target was appeared, the theta band of the frontal

midline did not increase the power .When the target was appeared on the screen, the increase of theta power appeared after one second in the second state. The theta band of the frontal midline increased the power until subjects almost find the target. The phenomenon was related with working memory because subjects needed to recall the location of the target by the landmarks. Current sensory input, spatial memory and motion are concurrently coded in the phase of every theta cycle to compute a possible direction of motion [69].

Figure 5-1: Time course of information processing in different layers [69].

Although subjects watched the landmarks in the first state, it seemed that subjects were less concerned with the landmarks and there was no variation of brain activity in the sensory information. In the previous task, the theta power returned to baseline state in the fourth state. In the present task, subjects were familiar with the virtual environment. The advantage of the simple structure kept off the influence from the structure of the virtual environment. We only considered the influence from different type of the landmarks, such as global and local landmarks. When the target was appeared, the weight of the landmarks was according to the target. The target affected the navigation strategies by which subjects decided the direction to go. In the second state, the phenomenon showed memory retrieval and the theta power increased.

In the third state, the phenomenon showed motion selection and the theta power still increased during navigation. In the fourth state, the theta power reduced to baseline.

We compared the responses of the brain activity in the global and the local conditions. The theta power of the frontal midline increased in both the global and the local conditions during spatial navigation. The increasing magnitude of the theta power was higher in the global condition than in the local condition.

We used permutation tests to determine the significant difference of the brain activities. There was significant difference between in the global condition and the local condition in comparison of theta band power in frontal midline. The marked gray region expressed the significant difference during spatial navigation. It was possible that the increase of the theta power in the frontal midline was linked to working memory processes. When subjects navigated to search the target in the virtual environment, subjects cognized the landmarks in the environment and recalled the relative location of target. It seemed that subjects played more attention for the global landmarks. Subjects acquired more information of the direction about guidance form the global landmarks.

VI. Conclusion

We designed a simple structure maze to test participants’ ability of navigation.

We could thus focus on the influence of different typical landmarks on navigation which was relatively independent from the geometric structures. We discussed several factors which may affect the behavioral performance in spatial navigation. We also reported results regarding how different types of landmarks influence spatial navigation.

Further, we investigated the brain dynamics associated with spatial navigation.

We adopted the independent component analysis (ICA) and event related spectral perturbations analysis (ERSP) to compare the EEG results corresponding to different types of landmarks. The theta power increased in the frontal midline component during spatial navigation. Females had better performance in the global condition than in the local condition in path ratio. The results of brain dynamics experiment consisted with the results of behavioral experiment. No matter subjects were familiar with the virtual environment or not, subjects spent more time on walking along paths that were not directly leading to the target. Performance differed between different of landmarks. We observed the significant difference in theta power between the global and the local condition by permutation tests.

Reference

[1] Amy L., Shelton, Timothy P. and McNamara, “Orientation and perspective dependence in route and survey learning,” Experimental Psychology Learning, Memory, and Cognition, 30, No. 1, 158–170, 2004

[2] Kuipers B., “Modeling spatial knowledge,” Cognitive Science, 2, 129–153, 1978 [3] Trullier O., Wiener S. I., Berthoz A. and Meyer J.A., “Biologically based artificial

navigation systems: review and prospects,” Progress in Neurobiology, 51, 483–544, 1997

[4] Sibylle D. Steck, Hanspeter A. and Mallot, “The role of global and local landmarks in virtual environment navigation,” Massachusetts Institute of Technology, 9, 1, 69–83, 2000

[5] Gillner S. and Mallot H., “Navigation and acquisition of spatial knowledge in a virtual maze,” Cognitive Neuroscience, 10, 445–463, 1998

[6] Witmer B. G., Bailey J. H. and Knerr B. W. and Parsons K. C, “Virtual spaces and real world places: Transfer of route knowledge,” Human Computer Studies, 45(4), 413–428, 1996

[7] Waller D., Hunt, E. and Knapp D., “The transfer of spatial knowledge in virtual environment training,” Presence: Teleoperator & Virtual Environments, 7(2), 129–143, 1998

[8] Maguire E. A., Frith C., Burgess N., Donnett J. G. and O’Keefe J., “Knowing where things are: Parahippocampal invement in encoding object locations in virtual large-scale space,” Cognitive Neuroscience, 10(1), 61–76, 1998

[9] O'Keefe J. and Nadel L., “The hippocampus as a cognitive map,” Oxford:

Clarendon, 1978

[10] Brian J. Stankiewicz and Amy A. Kalia, “Acquisition of structural versus object landmark knowledge,” Experimental Psychology: Human Perception and Performance, 33(2), 378-390, 2007

[11] Niedermeyer E., “The normal EEG of the waking adult,”

Electroencephalography, 149–173, 1999

[12] Raghavachari S., Rizzuto D. and Caplan J. et al., “Gating of human theta oscillations by a working memory task, ” J Neurosci , 21, 3175–3183, 2001 [13] Sederberg PB., Kahana MJ., Howard MW., Donner EJ. and Madsen JR., “Theta

and gamma oscillations during encoding predict subsequent recall,” J Neurosci 23, 10809–10814, 2003

[14] Walter F. Bischof and Pierre Boulanger, “Spatial navigation in virtual reality environments:An EEG analysis,” CYBERPSYCHOLOGY & BEHAVIOR, 6, 5, 2003

[15] Caplan J.B., Kahana M.J. and Sekuler R. et al., “Task dependence of human theta:

the case for multiple cognitive functions,” Neurocomputing, 32–33, 659–665, 2000

[16] Jeremy B. Caplan, Joseph R.Madsen and Andreas Schulze-Bonhage et al.,

“Human θ Oscillations Related to Sensorimotor Integration and Spatial Learning,” Neuroscience, 23(11), 4726–4736, 2003

[17] Kahana MJ., Sekuler R., Caplan JB., Kirschen MP. and Madsen JR., “Human theta oscillations exhibit task dependence during virtual maze navigation,”

Nature ,399 ,781–784, 1999

[18] Caplan JB., Madsen JR., Raghavachari S. and Kahana MJ., “Distinct patterns of brain oscillations underlie two basic parameters of human maze learning,” J Neurophys , 86 ,368–380. 2001

[19] Bland BH., “The physiology and pharmacology of hippocampal formation theta rhythms,” Prog Neurobiol, 26, 1-54, 1986

[20] Kamondi A., Acsady L., Wang X. and Buzsáki G., “Theta oscillations in

somata and dendrites of hippocampal pyramidal cells in vivo:

Activity-dependent phase-precession of action potentials,”Hippocampus, 8, 244-261, 1998

[21] Gevins A., Smith ME., McEvoy L. and Yu D., “High-resolution EEG mapping of cortical activation related to working memory: Effects of task difficulty, type of processing, and practice,” Cereb Cortex, 7, 374-385, 1997

[22] Burgess AP. and Gruzelier JH., “Short duration power changes in the

[22] Burgess AP. and Gruzelier JH., “Short duration power changes in the