Spatial Cognition


The research is focused on the navigation in the virtual tunnel task and its EEG correlates. We searched for the features in the EEG signal to discriminate the employment of the allocentric and the egocentric reference frames. These two reference frames differ in the center of deixis (the origin of the coordinating system). Part icipants tends to solve the task by adopting one of the mentioned reference frames. We decomposed the EEG signal to the basic features and used this data as the input for the neural networks. The classification task is to select the best features to discriminate between these reference frames. The result was congruent with the similar study (Gramann et al., 2006) in the Brodmann area 7 differences, but we also detected other brain areas involved in this task.

Fig. 1 Tunnel task. Participant has to traverse through the virtual tunnel consisting of 1st straigh segment, turned passage and the 2nd straght passage. At the end of the tunnel there are two arrows presented to the participant. His or her task is to decide which arror is pointing to the origin of the tunnel traverse.

Our research is focused on the representation of space and the employment of the reference frames. In the area of spatial cognition the reference frame is considered as orthogonal system with the origin (the deixis center) in the retina, head, body or other points, objects, or array in space. We presented modified version of the tunnel task (Gramann et al., 2006)  to our participants (see Fig.1)  to identify differences in the allocentric and egocentric reference frame processing. The difference between adoption of allocentric and egocentric reference frame is visualized in Fig.2.

Fig. 2 Example of egocentric and allocentric frame of reference adoption. At the beginning of the tunnel both frames are identical. At the end of the tunnel the egocentric frame is rotated the same angle as the head turns during the curved segment of the tunnel. The allocentric frame is fixed. The difference between these two angles is the same as the angle between the two arrows presented on the screen (dark bar) after the tunnel traverse.

The goal of the study is to localize brain areas involved in the processing of different frames of reference in 3D environment, so we presented tunnel travels in four directions (upward, downdard, left and right). We recorded the EEG signal during the traversing the tunnel. Experiment design consists of 20 traverses through the virtual tunnel. Subject has to choose after each traverse one of the two arrows pointing to the origin of the tunnel. The choice is the answer to the question, what reference frame he/she adopted as the navigation system.

We decomposed segments from each electrode to the signal features. There are 103 features for every electrode, inter-hemispheric and intra-hemisperic coherences of electrode pairs. Then we selected the best features discriminating between allocentric and egocentric reference frame. We adopted classical dual layer SOM network for the purpose of the analysis. The input layer contains 6 to 26 neurons according to the length of the input segment.

To distinguish the navigation in the horizontal and vertical direction we did the separate analysis just for the horizontal plane. There were some differences in both straight segments of the tunnel and the whole tunnel. The curved passage resulted in the same areas for the reference frame differentiation. The most frequent feature for all parts of tunnel was the coherence between the F8 and T4 electrode in the alpha band wave again.

Fig. 3. The comparison of the presented study (left) to the similar (Gramann, 2006).

We identify the differences in the processing of the allocentric and egocentric reference frame in the activity of the Broadmann area 7 in accordance to the similar study (Gramann et al., 2006). The processing in the area 7 is consistent with the neuroanatomic finding, because this area is considered as the center for spatial navigation and representation. The question is, whether we should attribute the differences in the processing of mentioned frames to the area 32 (Gramann at al., 2006). Moreover the distinction between results in horizontal plane and the both planes leads us to the conclusion there are special brain areas involved in the vertical navigation. We need to analyze these findings in more detail in the further research.


  1. Gramann, K., Müller, H., Schönebeck, B. & Debus, G. (2006). The neural basis of egocentric and allocentric reference frames in spatial navigation: Evidence from spatio-coupled current density reconstruction. Brain Research, 1118, 116-129.

Previous page: Emotional Processing
Next page: Projects in Biomedicine