In this study, we investigated the differences of brain activation between subjects using the allocentric reference frames and subjects using the egocentric reference frames during spatial navigation. Results showed that the navigation strategy do not have an apparent effect on behavioral performance in this simple tunnel task. The EEG results showed the use of allocentric and egocentric reference frame during navigation involved in different brain activations in the parietal and occipital regions.
Furthermore, the power changes at the frequencies around the alpha and beta bands were highly related the navigation performance, evaluated by errors of determining the homing direction, in the subjects using egocentric reference frame during passing the turn segment of the tunnel.
5.1 EEG dynamics associated with allocentric and egocentric representation
Results showed that many brain areas involved in spatial navigation processes and the phenomena of the consist activations in the parietal and occipital areas in during spatial navigation was plausible with their well-supported roles in visuospatial task. For example, the parietal lobe (Fig. 5-1) has known to play an important role in integrating sensory information coming from various parts of the body, particularly determining spatial sense and navigation (Blakemore and U. Frith, 2005). The occipital lobe which locates in the rearmost portion of the skull (Fig. 5-1) is the visual processing center of the human brain and is known can be broadly activated in visual tasks including color discrimination, motion perception, navigation, and even mental imagery (Kandel et al., 2000).
48
Figure 5-1: Location of the human brain lobes.
During tunnel passage, alpha suppressions associated with the path integration were found in both parietal and occipital area. Alpha suppression or event-related desynchronization (ERD) is well known for its relation to cortex activation (Pfurtscheller et al., 1996). The alpha suppression in occipital accompany with heading orientation may reflect activation of cortical to process the path integration visual flow information about rotation and translation during tunnel passage (Morrone et al., 2000; Kandel et al., 2000). The parallel alpha attenuation in the parietal area may involve in spatial awareness and processing of upcoming path information according to the dorsal pathway stretches from occipital lobe (Jong et al., 1994; Field et al., 2007). Subjects who used the allocentric reference frames showed stronger activation in occipital area comparison with the subjects using the egocentric reference frames. In contrast, subjects using the egocentric reference frames showed stronger activation in parietal during tunnel passage. The result is consistent with those neuron image studies which demonstrated that the tuners, using the egocentric reference frames, revealed stronger activation in parietal region during spatial navigation (Mellet et al., 2000; Galati et al., 2000; Committeri et al., 2004; Gramann
49
et al., 2006; Zaehle et al., 2007) whereas the subjects who used the allocentric reference frames activated a network comprising the parietal area and occipito-temporal area.
The alpha band power increased in parietal and decreased in occipital during deciding the homing direction and such alpha band power enhancement was thought to be an index of cortical inactivity (Pfurtscheller et al., 1996; Lin et al., 2005) or related to internally driven mental operations (Cooper et al., 2003, 2006; Klimesch et al., 2007). Since subjects had keep attention to the turning arrow and tried to recall the homing direction, the increased alpha power in the parietal area is not simply due to the cortical idling, but it may be an active processing which is necessary for internally commutating reference frames. Similarly, subjects who use an allocentric reference frame showed stronger alpha attenuation in occipital area during the homing direction selections. The task related alpha suppression in the occipital may not only reflect the process of visual information, but may also involve in the internal process of reference frame.
Results also revealed the different phasic power changes on EEG dynamics between allocentric and egocentric subject. The different tonic power changes on the theta, alpha, and beta band were showed in parietal area while only alpha band powers showed the different in occipital area between the two navigation strategy subjects.
The theta oscillatory has thought to encode the spatial representation. Specifically, Nishiyama and Yamaguchi (Nishiyama et al., 2002) showed that theta activity over frontal and parietal-temporal regions were associated with learning maze navigation and they also suggested that theta activity could reflect connections between hippocampus and frontal, parietal cortex. The significant differences on tonic power changes at several frequency band indicated that the parietal as well occipital regions
50
played different roles between two navigation strategy subjects.
In the left somatomotor ICs clusters, both allocentric and egocentric subjects revealed sustained alpha and its first harmonic suppression, also called mu suppression during tunnel passage and selecting homing angles. Mu rhythm is strongly suppressed during the performance of a motor action or intentional movements (Salmelin and Hari, 1994; Pfurtscheller et al., 1996; Pineda, 2005).
During tunnel passage, subject could percept the perspective visual flow of motion and then required to press button for answering direction and therefore the mu blocking showed in this area was expectable.
5.2 EEG Dynamics associated with navigation performance
Relationship between the brain activity and navigation performance was first demonstrated in this study. Comparing to trails with well-estimation of homing angle, parietal EEG activities of subject who use egocentric representation at the frequencies from 8 to 30 Hz were consistently stronger attenuated when subject overestimated homing angle. Oppositely, when subject underestimated homing angle, the EEG power at the same frequencies range were consistently less attenuated. The egocentric representation has suggested to involve the dorsal pathway (Sdoia et al., 2004;
Committeri et al., 2004) which has also been reported relate to the accuracy of the predictions but not the content of subject’s perceptual (Donner et al., 2007). Other researchers also found accuracy-predicted pre-stimulus brain activity (Makeig and Jung, 1996). Therefore this 8-30 EEG activity may be an index of navigation performance for egocentric subjects.
The hippocampus provides an allocentric representation of space and has found to associate with navigation performance. The higher accuracy of navigation was measured, the stronger activation of the right hippocampus was observed (Maguire et
51
al., 1998). Patients with hippocampal lesions also showed poor performance in large-scale complex environment navigation (Maguire et al., 2006). Brain activity in hippocampus is almost impossible to be detected from the scalp EEG due to its deep location in brain. Therefore, even though subjects who use an allocentric representation thought to involve visual -temporal pathway (Sdoia et al., 2004;
Committeri et al., 2004), it is probably the reasons for explaining why we couldn’t observe the specific performance related brain dynamics among occipital and temporal regions in the allocentric subjects.
52