VIII. 10. Right Temporal-Lobe Contribution to the Retrieval of Family Relationships in Person Identification

Transcription

1 CYRIC Annual Report 2009 VIII. 10. Right Temporal-Lobe Contribution to the Retrieval of Family Relationships in Person Identification Abe N. 1, Fujii T. 1, Ueno A. 1, Shigemune Y. 1, Suzuki M. 2, Tashiro M. 2, and Mori E. 1 1 Department of Behavioral Neurology and Cognitive Neuroscience, Tohoku University Graduate School of Medicine 2 Division of Cyclotron Nuclear Medicine, Cyclotron and Radioisotope Center, Tohoku University We previously reported a patient who exhibited a peculiar form of delusional misidentification 1). She had a selective deficit in retrieving family relationships between herself and her daughters and husband (i.e., she misidentified her daughters as her sisters and her husband as her father) despite being able to retrieve their names and faces and some person-specific semantic information (e.g., occupation). Based on this finding, the present positron emission tomography study (PET) was designed to elucidate the brain mechanisms underlying the retrieval of family relationships in healthy individuals. Fifteen male volunteers who had no history of neurological or psychiatric disease were paid for their participation in this study. Before the analysis of imaging data, three participants were excluded due to excessive head motion during PET scanning (approximately 5 mm). Thus, the results of the present study are based on the data from the remaining 12 subjects (age range years, mean age 21.6 years). There were no pathological findings during magnetic resonance imaging (MRI) of any of the subjects brains. All of the subjects were right-handed. The subjects gave their written informed consent in accordance with the Declaration of Helsinki and the guidelines were approved by the Ethical Committee of Tohoku University. For each subject, we collected 6 photographs of the faces of the subject s family members. These 6 photographs consisted of different family members, and each photograph of a face represented one of the following: mother, father, grandmother, grandfather, elder brother, younger brother, elder sister, younger sister, aunt, uncle, niece, and nephew. All of the photographs had been taken within the past few years. During PET 187

2 imaging, the subjects performed three tasks: (1) a name task in which subjects were presented with photographs of the faces of family members and were asked to retrieve names and answer them orally; (2) an occupation task in which subjects were presented with photographs of the faces of family members and were asked to retrieve occupations and answer them orally; and (3) a relation task in which subjects were presented with photographs of the faces of family members and were asked to retrieve family relationships and answer them orally. In addition to these three tasks, subjects also performed two control tasks, a judgment of the gender of family members in one task and of unfamiliar people in the other task. Regional cerebral blood flow (rcbf) was measured using PET (SET2400W Shimadzu, FWHM 4.0 mm) and 15 O-labeled water (approximately 180 MBq per injection). The transaxial sampling field of view (FOV) was 256 mm, and the axial FOV was 190 mm. The thickness of the slices measured was mm. Before the PET experiments, a catheter was inserted into the subject s right brachial vein for tracer administration, and the subject s head was fixed in an air-cushioned headrest apparatus. Each task started 1.5 s before PET data acquisition, and PET data acquisition lasted 60 s. A transmission scan was followed by the experiment, and the data were used to obtain corrected emission images. The PET data were analyzed with SPM8 (Wellcome Department of Imaging Neuroscience, London, UK). The threshold of significance was set at p<0.001 (uncorrected for multiple comparisons) with an extent threshold of 10 contiguous voxels. Behaviorally, all of the subjects were 100% correct in each task; they could completely retrieve names, occupations, and relations in response to each stimulus. As for the imaging data, to identify the neural activity specifically associated with the retrieval of names of family members, we performed a conjunction analysis of name vs. occupation and name vs. relation. We did not find a significant difference in the neural activity of any brain region. Then, to identify the neural activity specifically associated with the retrieval of the occupations of family members, we performed a conjunction analysis of occupation vs. name and occupation vs. relation. This analysis detected a significant activation in the left middle temporal gyrus. Next, to identify the neural activity specifically associated with the retrieval of family relationships, we performed a conjunction analysis of relation vs. name and relation vs. occupation. This analysis detected a significant activation in the right inferior temporal gyrus. These results are summarized in Table 1 and Fig. 1. In the present study, we used PET to identify the neural correlates of the retrieval of 188

3 family relationships in person identification. We found that the retrieval of family relationships was specifically associated with activity in the right inferior temporal gyrus. The right inferior temporal gyrus was one of the regions showing prominent hypoperfusion in our previously reported patient who showed a selective deficit in the retrieval of family relationships in person identification 1). The present findings, together with our case report, support the view that the retrieval of the relationships between oneself and family members may require a special cognitive process dissociated from the retrieval of names and other person-specific semantic information. Most importantly, we found that right inferior temporal activation was specifically associated with the retrieval of family relationships. This finding is partially consistent with our a priori hypothesis that the activation relevant to the retrieval of family relationships could be found in the right fronto-temporal region. However, we did not find right frontal activation in this study, suggesting the possibility that the right frontal lobe is not associated with the retrieval of family relationships per se. It is difficult to say what roles the right inferior temporal gyrus plays in the retrieval of family relationships, but we should note the special nature of the conceptual knowledge of family relationships. Basically, knowledge about family relationships is relative semantic information determined by the relationship between oneself and a partner, whereas other knowledge, including names, faces, and some person-specific information (e.g., occupation), is absolute semantic information, which does not depend on the relationship between oneself and a partner. Thus, the process of retrieving family relationships between oneself and one s partner is likely to require an additional cognitive process that monitors relative semantic information, which might be associated with the right inferior temporal gyrus. Together with the findings of our case study 1), the present imaging findings support the view that the retrieval of the relationships between oneself and family members may require a special cognitive process dissociated from the retrieval of names and other person-specific semantic information. The present study also suggests that the right inferior temporal gyrus may play an important role in the retrieval of family relationships in person identification. One major limitation in the current study is that we could not dissociate the brain activations in response to highly intimate family members (e.g., mother and father) and less intimate family members (e.g., aunt and uncle). The effect of familiarity feeling on brain activations during the retrieval of relations is an important issue to be pursued. Psychological data regarding the subjective feeling of familiarity to family members, 189

4 combined with an event-related design using fmri, could have been informative. It should be also noted that the present findings can be specified only in semantic representations of family members and cannot be generalized in person-related semantics including famous or newly learned people, since the family relationships for an individual exist only between the individual and his/her family members. Although more data collection concerning the neural mechanisms underlying the retrieval of family relationships (e.g., neuropsychological study of sizable series of patients) is required, the present findings provide further insight into the neural mechanisms underlying person identification and theoretically stimulate the updating of neurobiologically-based models of person identification. Reference 1) Abe N., Neuropsychologia 45 (2007) Table 1. Brain regions showing specific responses across name occupation and relation tasks. Regions (Brodmann s area) Coordinates x y z Z value Cluster size Conjunction analysis of name vs. occupation and name vs. relation No suprathreshold activations Conjunction analysis of occupation vs. name and occupation vs. relation Left middle temporal gyrus (21) Conjunction analysis of relation vs. name and relation vs. occupation Right inferior temporal gyrus (20)

5 Figure 1. Significant activations detected in SPM whole-brain analyses. The activations are superimposed onto the standard brain. The rcbf values of the following two regions are depicted (error bars represent S.E.M.). (A) The left middle temporal gyrus is activated during the retrieval of occupation compared with the retrieval of name and relation. (B) The right inferior temporal gyrus is activated during the retrieval of relation compared with the retrieval of name and occupation. 191