Thinking of mental and other representations: The roles of left and right temporo-parietal junction

Abstract

Adopting new versions of the false belief and “false” photo vignettes used by Saxe & Kanwisher (2003) we were able to show activation in all five areas previously reported. Activations by added false sign and temporal change control vignettes in these areas showed that the right temporo-parietal junction (TPJ-R) is specifically associated with processing mental states like belief. In contrast, TPJ-L was also activated by false signs suggesting an association with processing perspective differences for mental and non-mental objects in line with work on visual perspective tasks. A similar, but less clearly interpretable pattern of activations was also found in the middle temporal gyrus, precuneus and in the MPFC. These results are discussed in relation to findings in normal development where false belief and false sign tasks associate more strongly with each other than does each of them with the “false” photo task.

Notes

¹This difference can also be brought out more clearly by making the difference in target explicit in the content:

At time 2, the observer thinks the lady is wearing red at time 2.

At time 2, the photo shows the lady wearing red at time 1.

At time 1, the lady was wearing red.

At time 2, the lady is wearing green.

Since only the belief's content is at odds with its target (lady's costume at time 2) but not the photo's (it is about time 1), it is only the belief that is false but not the photo.

²Parkin also tried to implement genuinely false photos by taking photos of, e.g., yellow objects through a blue filter making them look green in the photo. However, making children understand what the filter did to the photo was a difficult methodological problem.

³We should point out that when using a different criterion of significance of p<.05 corrected at cluster level then PH activates significantly more strongly in this region than TCC (p<.019).

⁴This requires some justification. Vogeley et al. (2004) reported activation of TPJ-L only in their 1st person condition over their 3rd person condition, when one would expect perspective differences to become relevant only in the 3rd person condition. However, Aichhorn et al. (2006) argued that the specific task used by Vogeley et al. makes it likely that participants were more likely to become concerned about perspective differences in the 1st person condition, because they may have wondered how many objects there really were as opposed to how many they were made aware of. Whereas the 3rd person condition participants may simply have focused on comparing how many target objects there were (they could see) with how many of these were in front of the avatar.