KJ00004432262.Pdf

Joumal Joumal of Hiroshirna Prefectural College of Health Sciences Hurnanity and Science 5 (1) 21- 30 2005

Organization Organization and Development of the Brain Mechanism for for Understanding Speakers' Reallntentions

Satoshi Satoshi IMAIZUMI*1 Midori HOMMA 勺 Y oshiaki OZA W A* 1

Kazuko Y AMASAKI 叶

*1 Hiroshima Prefectural College of Health Sciences ，Japan *2 Hiroshima Prefectural Rehabilitation Center ，Hiroshima ，Japan

Abstract Abstract

To clarify how the brain understands the speaker' s rnind for verbal acts ，fM RI irnages obtained frorn 24 subjects and behavioral behavioral data obtained frorn 339 subjects were analyzed when they judged the linguistic rneanings or emotional rnanners of of spoken phrases. The target phrases had linguistically positive or negative rneanings and were uttered warrnheartedly or coldheartedly coldheartedly by a wornan speaker. The results of the 品1RI analyses suggest that neural resources responsible for the speakers' speakers' mind reading are distributed over the superior ternporal sulci (STS) ，inferior frontal regions ，rnedial frontal regions and and posterior cerebellum. When cornpared to the female subjects ， the rnale subjects showed significantly stronger activation only only in the right frontornedian cortex ， which can be hypothesized to infer the speaker' s hidden but true intentions frorn the speaking speaking acts. The correct judgrnent of the speaker's intentions significantly increased with age for the phrases with inconsistent inconsistent linguistic and emotional valences. Female children showed faster developrnent than rnale children. These results suggest suggest that neural processes for language and ernotional prosody significantly interact to rnentalize verbal acts ， and sex differences differences exist in this rnind reading system

Key words :rnind reading for verbal acts ，brain ，developrnent ， speech cornrnunication

円/“噌Bi Journal Journal of Hiroshima Prefectural College of Health Sciences Humanity and Science 5 (1) 21 - 30 2005

surveyed. surveyed. INTRODUCTION Using Using spoken phrases with positive or negative linguistic Phrases Phrases with positive linguistic meanings may convey meanings uttered by a female with two emotions ， negative negative meanings when uttered with coldhearted emotion. warmhearted and coldhearted ，we analyzed brain activities For instance ，“It' s wonderful" uttered with evident sarcasm based on a functional MRI measurement when subjects may convey that “It' s not wonderful at all." On the judged linguistic and emotional meanings of the phrases ， contrary ， phrases with negative linguistic meanings may and discussed possible neural mechanisms for mentalizing convey positive meanings when uttered with warmhearted speaking acts. We also analyzed the development of the emotion. “How stupid you are" may be interpreted as “How mentalizing ability ，testing young people aged 6 to 28 years nice nice you are" when uttered with warmhearted emotion and of age. a friendly feeling. Listeners have to correctly interpret a speaker' speaker' s emotion and integrate it with the linguistic METHOD meaning of the speech act to understand the hidden but true intentions intentions of the speake r. Little is known about how All experiments in this study were conducted in linguistic linguistic and emotional information is integrated in the accordance with the De c1 aration of Human Rights ，Helsinki brain ，and how this brain mechanism develops and when it 1975 and the research ethics regulations in the authors' matures. matures. affiliation. Written informed consent was obtained from Humans strongly and f1 exibly interpret the speaking each subject after explaining the pu 中 ose and the outline of behavior behavior of other people based on an understanding in their the method for this research and the advantages and own minds. The awareness that other people have beliefs disadvantages expected for the subjects. and desires different from our own and that their speaking behavior behavior can be explained by these beliefs and desires can Speech materials

be be referred to as the “ theory of mind" 1， 2) or “mentalizing" 3). Two sets of speech samples were prepared: one for Most conventional theory of mind tests based on a false testing young children ，and the other for testing adults. For belief belief estimate that the “mentalizing" ability matures by as testing adults ， frequently used 40 phrases with positive early early as age four. linguistic meanings ， such as “1 love you ，" and 40 phrases Based on the homologs between studies of single-cell with negative linguistic meanings ， such as “1 hate you ，" activity activity in monkeys and areas implicated in mentalizing were uttered by a female speaker with a Tokyo dialect ， with from brain imaging studies ， Frith & Frith (1999)3) positive and negative emotions ， that is ， warmhearted and suggested suggested that the components of the mentalizing system coldhearted emotions (Table 1). The warmhearted include include the (i) superior temporal sulci (STS) ， for the utterances were made showing strong pleasure ， while the detection detection of the behavior of agents and analysis of the goals coldhearted ones were made showing strong hatred. These and outcomes of this behavior; (ii) inferior frontal regions ， utterances were c1 assified into four types: “praise" as the for for the representations of actions and goals; and (i ii) linguistically and emotionally positive phrases ，“ irony" as anterior anterior cingulate cortex / frontomedian regions ， for the the linguistically positive and emotionally negative phrases ， representations representations of mental states of the self. An implication “ banter" as the linguistically negative and emotionally of of their proposal is that the analysis of another agen t' s positive phrases ，and “blame" as the linguistically and behavior behavior in co 吋unction with the representation of our own emotionally negative phrases. These utterances were mental mental states allows us to make inferences about the digitized and analyzed to extract prosodic characteristics intentions intentions of that agen t. Although such a mentalizing such as FO contour and utterance length ，and then used for system system might contribute to the interpretation of speaking listening tests during functional brain imaging. acts ，no systematic study has been carried out as far as we For testing young children ，frequently used 20 phrases

Table 1. Speech materials used for the experiments.

¥Lin¥EjuZ 川m町oMMoneaal nMi 白官ar¥m¥ er Warmhearted (+) Coldhearted ←) Total

Positive Positive (+) 40 (praise) 40 (irony) 80 Negative Negative (ー) 40 (banter) 40 (blame) 80 Total Total 80 80 160

-22- Joumal Joumal of Hiroshirna Prefectural College of Health Sciences Hurnanity and Science 5 (1) 21 - 30 2005

J Y o J y o f

400 400 、 :--d b !j !j 300 医主 、 Z ~ 200j 位4 ヨ ーー----一一回 d c 100 100 0.0 1. 0 Time (s)

Figure Figure 1. FO contour of two utterances of /iiyo/ (rneaning “ okay") with warrnhearted ernotion (l eft) and and with coldhearted ernotion (right). a: initial FO ，b: rnaximum FO ，c: minimum FO ，d: final FO ， FO range= b-c ，length=time difference between d and a. with with positive linguistic rneanings ， such as “How nice you whether the pitch of a tone rises or falls. are!" are!" and 20 phrases with negative linguistic rneanings ， With the subject in the fM RI systern (Siernens ， Magnetic such such as “How bad you are!" were uttered by a female Symphony 1.5T) ，one of the listening tasks and the control speaker speaker with a Tokyo dialect with positive and negative task were altemately perforrned at 30 ・second intervals ，4 ernotions. ernotions. These utterances were digitized and analyzed to tirnes each ， and this block design ，which took 4 rninutes to extract extract prosodic characteristics ， and then used for listening cornplete ，was perforrned for the 2 listening tasks. The tests. tests. speech and tone stirnuli were presented at the rnost comfortable comfortable level for the individual subjects ， once per 3 Acoustic Acoustic analyses seconds. Pressing two buttons with the right index and

As shown in Figure 1， fundamental frequency at various rniddle fingers ， the judgment and the 陀 sponse tirne were characteristic characteristic timing points and total length of utterances rneasured. were measured and analyzed by ANOV A with two factors The scanning conditions were: the interscan interval consisting consisting of ernotional Manner (Warrnhearted vs. (TR) ，6 seconds; acquisition tirne (TA) ，4 .4 seconds; flip

0 Coldhearted) Coldhearted) as a within-phrase rneasure and linguistic angle ，90 • The irnage data obtained were analyzed using a Meaning (Positive vs. Negative) as a between-phrase statistical pararnetric mapping (SPM2 ， the Wellcorne measure. measure. Department of Imaging Neuroscience ，http://www.fi l. ion.

uc l. ac.u k/ sp m/ spm2.htrnl) 4). Li stening tasks for MRI measurement The listening subjects for the brain imaging were 24 Developmental analyses

right 置 handed healthy adults (1 2 males and 12 fernales; In order to develop a screening test for the early average average age 24.7 years). Two listening tasks were detection of communication problems in children ， the irnposed ， the language task and the emotion tas k. For the rnind-reading ability from speech was rneasured for 339 language language task ， the listening subjects were instructed to listening subjects (1 73 male and 166 female) using spoken judge judge quickly whether the linguistic rneaning of a presented phrases directed to children based on the two tasks ， the speech speech sarnple is positive or negative. For the emotion task ， language task and the ernotion tas k. Table 11 shows the age they they were instructed to judge as quickly and correctly as and number of subjects for the developrnental analyses. possible possible whether the emotional prosody of a presented speech speech sarnple is warrnhearted or coldhearted. RESULTS The response tirne and the percent correct were analyzed by ANOV A with three factors ，including emotional Manner Acoustic characteristics of speech materials (Warrnhearted (Warrnhearted vs. Coldhearted) ， linguistic Meaning Acoustic characteristics varied depending on Manner and (Positive (Positive vs. Negative) ， and Sex (Fernale vs. Male). Meaning. As illustrated in Figure 2，ANOV A on the FO For For the control task used in brain irnaging ， tones with a range revealed that the interaction effect of Manner and rising rising pitch or a fa l1i ng pitch were prepared. The length and Meaning (F (1， 160)=4.97 ，p=0.027) ， the rnain effects of the the presentation level of the tones were adjusted ， as well as Manner (F (l， 160)=63.21 ， Pく0.000 1) and Meaning those those of the speech sarnples. The control task was to judge (F (l， 160)=9.98 ，P く0.005) were significan t. The post-hoc

内ノ臼円〈 U Journal Journal of Hiroshima Prefectural College of Health Sciences Humanity and Science 5 (1) 21 - 30 2005

Table Table 11. Subjects for developmental analyses Stage Stage Age Male Female Total MeanAge 7 14 15 29 Primary Primary 20 14 34 7.7 School School 8 9 19 16 35 10 10 19 18 37 Middle Middle 18 18 17 35 10.5 School School 11 12 12 20 19 39 13 13 18 18 36 High 17 17 20 37 13 .4 School School 14 15 15 17 15 32 Adu 1t 20-28 10 15 25 24.1 total total 173 166 339

30 日 1.2

250 250 (品巴 戸---L一一 upH 事 B 帽凶 ← .8 CH S 円 tl tl b ，、 50 50 .ι

不「白百 iliearted Coldhearted 市rarmhe 訂 ted Coldhearted Emotion Emotion

Figure Figure 2. (a ，left) FO range with error bars. (b ，right) Utterance length with error bars. Spoken phrases phrases directed to adu 1t s. Empty bar: Negative linguistic meanings ，Meshed bar: Positive linguistic meanings.

test test revealed that the warmhearted utterances had 104Hz rnore accurately judged than those of the positive phrases ， wider wider FO range than the coldhearted utterances (p

q/A 』 斗 A Joumal Joumal of Hiroshima Prefectural College of Health Sciences Humanity and Science 5 (1) 21- 30 2005

Table 111. Percent correct of the two 岱 IRI tasks. Task Task Sex Praise(++) Irony(+ 一) Banter( -+ ) Blame( -- ) Emotion Emotion Female 88 95 73 99 Emotion Emotion Male 84 84 61 91 Language Language F emale 96 87 87 86 Language Language Male 96 78 84 93

Table IV. Response time measured under the two fM RI tasks. Task Task Sex Praise(++) lrony(+ ー) Banter(-+) Blame(--) Emotion Emotion Female 1.2 0 1.30 1.2 5 1.17 Emotion Emotion Male 1.5 5 1.76 1.67 1.54 Language Language Female 1.2 1 1.48 1.31 1.32 Language Language Male 1.4 1 1.71 1.61 1.61

for for coldhearted utterances of negative phrases than those of cerebellar lobe was the only area ， the activity of which was positive positive phrases. The main effect of Sex was also significant at PPWE-corr く0.05 corrected for multiple significant significant (F (1， 72)=98.7 ，p

Brain Brain imaging results As shown in Figure 3， the male subjects showed significant significant activation in more cortical areas than the female subjects subjects for both tasks. For the male subjects ，significant activation activation (PPWE-corr<0.05) was found in the bilateral (b) Male brain middle middle temporal gyri ，including the superior temporal sulci ， the the bilateral superior frontal gyri ， the left posterior Figure 3.Brain activity pattem revealed in the emotion task- the control task (p

円ノム】Fhu Joumal of Hiroshima Prefectural College of Health Sciences Humanity and Science 5 (1) 21-30 2005

Fem 誠e subjects Male :ru bjects

100 100 100

〈 JHmU (器官むと 芯 ω-H-MOU 30 80 GOES 制CE 60 60 60 包内問 仏

40 40 C h.i1 d Low Te ens High Te ens Ad u1 ts C h.i1 d Low Te ens High Te ens Ad u1 ts 勾e Ag e Figure Figure 4. Development of the mind-reading ability. • praise (l anguage +， emotion +)， 0 irony (l anguage +， emotion -)， • banter (l anguage -， emotion +)，ム blame (l anguage -， emotion -).

correctness correctness scores than teenage female subjects in junior Other results suggest that men and women show very

and and senior high schools. The sex difference ，however ，was similar ，strongly left lateralized activation pattems 6). Thus ， not not significant for the adult subjects. the debate on sex differences in the functional organization of of the brain for language and prosody has not yet been resolved. resolved. DISCUSSION As shown in the behavioral data ， the male subjects When compared to the female subjects ， the male subjects showed lower correctness scores with longer RT than the showed significantly stronger activation in only one cortical female subjects. One possible interpretation of the present area area in the right frontomedian cortex only under the results is that the male subjects need conscious inferences emotion emotion task. The frontomedian cortex has been shown to about the verbal acts and intentions of the female speaker ，

be be important for theory 自 of-mind processes ，i.e. ， during the which required stronger activation of relevant neural attribution attribution of other people's actions to their motivations ， substrates with longer RT than the female subjects. Ferstl

beliefs ， or emotions 2). It has also been suggested that the and Cramon (2002 )2) suggest that the essential role of the right-hemisphere-damaged right-hemisphere-damaged but not the left-hemisphere- frontomedian area is inference ，which is a core component damaged patients have difficulties in theory-of-mind for the theory of mind processes. The female subjects may

processes processes 1). Therefore ， the present results may suggest that need less conscious inferences to process emotional the the right frontomedian area of the male subjects contribute prosody. These results may coincide ， in part ， with previous to to analyzing the behavior speaking of the female speaker in studies suggesting that emotional prosody modulates conjunction conjunction with the representation of their own mental perceptual word processing and that the time-course of this states ，which allows them to make inferences about the modulation differs for males and females; that is ，women intentions intentions of the female speake r. make an earlier use of emotional prosody during word

A question arises as to why the female subjects do not processing as compared to men 9l ; and that women ， but not need need activation in this area. A much debated question is men ，show an interaction between prosody and word

whether whether sex differences exist in the functional organization valence during a semantic processing stage lO l. If so ，then ，

of of the brain for language and prosodys- I2). A long-held men ， not women ， would need to integrate linguistic hypothesis hypothesis posits that language functions are more likely to semantics and emotional intent at a higher stage than the be be highly lateralized in males and to be represented in both semantic processing stage ，and therefore men may need cerebral cerebral hemispheres in females ， but attempts to extra activations of relevant neural resources and eventually demonstrate demonstrate this have been inconclusive. Some data need longer RT than women. indicate indicate that language primarily lateralizes to the left in The male subjects in the present study show significant males; males; and approximately half have left lateralization ， and activation in the cortical areas including the bilateral

the the other half have bilateral representation in females I2). superior temporal sulci (STS) ， the left inferior frontal

q/nhu

ム】 Journal Journal of Hiroshirna Prefectural College of Health Sciences Hurnanity and Science 5 (1) 21 - 30 2005

region ， the medial frontal region ， and the posterior their goals. cerebellum ， while the female subjects show only non- While some data indicate that both emotional and significant significant weak activations in these areas ， except for the linguistic prosody are processed in the right hemisphere ， posterior posterior cerebellum. These results suggest that the mind other results suggest that only emotional prosody or only reading reading system for speaking acts consists of neural linguistic prosody is processed in the right hemisphere. resources resources distributed over these areas ， and that there are Sorne results imply that sentence-levellinguistic prosody is sex-dependent sex-dependent differences in this mind system. reading processed in the left hemisphere. Alternatively ，it has been Some of these cortical areas may be in connection with proposed that the lateralization of prosodic processing may theory-of-mind theory-of-mind processes L2 ，13.14.15l or mentalizing vary as a function of the acoustic parameters of prosody ， processes 3l ，i.e. during the attribution of other people's such as fundamental frequency ，intensity or duration under actions actions to their motivations ，beliefs ， or emotions. Frith & stud y22 l. Some others suggest that ths lateralization of Frith Frith (1 999)3 1， for instance ， proposed a distributed neural prosodic and phonetic processing may vary depending on menta 1i zing system based on the homologs between studies the subjective attention 23 .24 ). It has also been suggested that of of single-cell activity in monkeys and areas implicated in emotional prosody modulates word processing and that the mentalizing mentalizing from brain imaging studies. It consists of time-course of this modulation differs for males and (i) (i) STS ， for the detection of the behavior of agents and females ， that is ，women make an earlier use of emotional

9 1O analysis analysis of the goals and outcomes of this behavior; prosody during word processing as compared to men • ). (ii) (ii) inferior frontal regions ， for the representations of Although several previous studies have suggested actions actions and goals; and (iii) anterior cingulate cortex / otherwise ， neither the right-hemisphere predominance in frontomedian frontomedian regions ， for the representations of mental the emotion task nor the left-hemisphere predominance in states states of the sel f. The present study suggests that rnind- the language task was observed in this study. One possible reading reading for verbal acts relies also on the posterior interpretation of the present results is that the linguistic as cerebellum cerebellum cooperating with the above mentioned well as emotional processes may automatically be induced mentalizing mentalizing system. as cooperative and interactive processes ， and the The activity in the posterior cerebellum was found for differences may exist only at a high processing stage to both both sexes under both tasks ， although the location of the integrate the linguistic and emotional processes to generate

maximum activity varied depending on the sex and task. a coherent interpretation of the speaker's mind25 • 26ヱ 7) The maxirnum activity for the males was located in the left As shown in Figure 2，significant interaction effects of posterior posterior cerebellum under the emotion task and in the right language and emotion were observed on the acoustic under under the language task ， while it was located in the right characteristics of utterances ， such as the FO range ，and also posterior posterior cerebellum for the females under the emotion on the perceptual behavior evaluated by response time and task. task. This activation suggests a role in the processing of judgment correctness. These results obtained in our language language and emotion for the posterior cerebellum. experiment suggest that emotion modulates linguistic Evidence Evidence has accumulated suggesting that the cerebellum processes not only in speech production but also in speech contributes contributes to cognitive functions not only linguistic perception ， and such modulations may differ between the semantic semantic processing l6l ， but also affect processing and sexes ，particularly in perceptual processes. These results mentalizing l7l ， and empathy with mirroring for emotive support ， in part ， previous studies suggesting that emotional

actions l8l . Lesion studies l9l suggest that patients with prosody modulates perceptual word processing and that the lesions lesions involving the posterior cerebellum are characterized time-course of this modulation differs for males and by impairment of verbal executive functions such as verbal females ， that is ，women make an earlier use of emotional

fluency ，personality change with blunting of affect ， and prosody during word processing as compared to men9.lO l• language language deficits ，in c1 uding agrammatism and dysprosodia. The correctness score ， as shown in Figures 4 and 6， Although Although the primary role of the cerebellum has been significantly increased with age for irony and banter considered considered to have phrases

内ノ円 i' L】 Joumal Joumal of Hiroshima Prefectural College of Health Sciences Humanity and Science 5 (1) 21 - 30 2005

by the conventional theory of mind tests based on a false role of emotional prosody during word processing. belie f. This also suggests that the emotion task developed Brain Res Cogn Brain Res ，14(2):228-233 ， 2002. here here is useful to measure the speech communication ability ， 11) Schirmer ， A. and Kotz ， S.A.: ERP evidence for a sex- or or the ability to understand speakers' minds for verbal acts. specific stroop effect in emotional speech. J Cognitive Neuroscience ，15(8): 1135-1148 ， 2003. 12) 12) Shaywitz ，B.A. ，Shaywitz ，S.E. ， et a1.: Sex differences CONCLUSION in in the functional organization of the brain for language. The present results suggest that neural resources Nature ，373(6515):607-609 ， 1995. responsible responsible for the mind reading of speech are distributed 13) Vikingstad ， E.M. ， George ， K.P. ， et al.: Cortical over over the STS ，inferior frontal regions ， medial frontal language lateralization in right handed normal subjects regions regions and posterior cerebellum. The neural processes for using functional magnetic resonance imaging. J Neurol language language and emotional prosody significantly interact to Sci ，175 (1): 17-27 ， 2000. mentalize mentalize verbal acts. The neural mechanism to inte 中ret a 14) Premack ， D. and Woodruff ， G.: Chimpanzee problem- speaker' speaker' s real intentions from spoken phrases with solving: a test for comprehension Science ，202 ，532- inconsistent inconsistent linguistic and emotional valences develops 535 ， 1978a. rather rather slowly during the junior high school stage ， although 15) Premack ，D. ， and Woodruff ， G.: Does the chimpanzee females females develop faster than males. have a theory of minds? The Behavioral and Brain Science ，1 ，515-526 ， 1978b. 16) 16) Baron-Cohen ，S. ，Leslie ， A. M. ， et a1.: Does the autistic REFERENCES child child have a “theory of mind"? Cognition ， 21 (1)， 37- 1) 1) Siegal ，M. ，Carrington ，J. ， et a1.: Theory of mind and 46 ， 1985.

pragmatic pragmatic understanding following right hemisphere 17) Roskies ， A. L.， Fiez ， J. A. ， et al.: Task 国 dependent damage. damage. Brain Lang ，53(1):40-50 ， 1996 modulation of regions in the left inferior frontal cortex 2) 2) Ferstl ， E.C. and von Cramon ， D.Y.: What does the during semantic processing. Cognitive Neuroscience ， frontomedian frontomedian cortex contribute to language processing: 13(6) ，829-843 ，200 1. coherence coherence or theory of mind? Neuroimage ，17(3): 1599- 18) Gundel ， H. ， O'Connor ， M. F. ， et a1.: Functional 612 ， 2002. neuroanatomy of grief: an FMRI study. American 3) 3) Frith ， C. D. and Frith U.: Interacting Minds -A Joumal ofPsychiatry ，160 ，1946-1953 ， 2003. Biological Biological Basis. Science ， 286: 1692-1695 ， 1999. 19) Leslie ，K. R ，Johnson-Frey ， S. H. ， et a1.: Functional 4) 4) Statistical Parametric Mapping: SPM2 ， Developed imaging of face and hand imitation: towards a motor by members & collaborators of the Wellcome theory of empathy. Neuroimage ，21(2) ，601-607 ， 2004. Department Department of Imaging Neuroscience 20) Schmahmann ，J. D. ， and Sherman ，J. c.: The cerebellar http://www.fi 1. ion.uc 1. ac .u k/ sprnlspm2.html cognitive affective syndrome. Brain ， 121 ( Pt 4) ，561- 5) 5) Baxter ， L.c.， Saykin ，A.J. ， et a1.: Sex differences in 579 ， 1998. semantic semantic language processing: a functional MRI study. 2 1) Imamizu ，H. ，Miyauchi ，S. ， et a1.: Human cerebellar Brain Brain Lang ，84(2):264-72 ， 2003. activity reflecting an acquired intemal model of a new 6) 6) Frost ，J.A. ，Binder ，J. R.， et al.: Language processing is too 1. Nature ，403(6766) ，192-195 ， 2000.

strongly strongly left lateralized in both sexes. Evidence 仕om 22) Imamizu ，H. ，Kuroda ，T. ， et a1.: Modular organization

functional functional MR I. Br 泊n， 122 (Pt 2):199-208 ， 1999. of intemal models of tools in the human cerebellum. 7) 7) Koelsch ，S. ，Maess ，B. ， et a1.: Electric brain responses Proc Natl Acad Sci U SA ，100(9) ，5461-5466 ， 2003. reveal reveal gender differences in music processing. 23) Zatorre ， R. J. ， Belin ， P. ， et al.: Structure and Neuroreport ，14(5):709-13 ， 2003. lateralization of auditory cortex: Music and speech. 8) 8) Kotz ，S.A. ，Meyer ，M. ， et al.: On the lateralization of Trends in Cognitive Science ，1(6) ，37-46 ， 2002. emotional emotional prosody: an event-related functional MR 24) Imaizumi ，S. ，Mori ， K.， et al.: Vocal identification of

investigation. investigation. Brain Lang ，86(3):366 由 376 ，2003. speaker and emotion activates different brain regions. 9) 9) Morris ，1. S. ，Scotto ，S. K.， et a1.: Saying it with feeling: NeuroReport ，8(5) ，2809-2812 ， 1997. neural neural responses to emotional vocalizations. 25) Imaizumi ，S. ，Mori ， K.， et a1.: Task-dependent laterality Neuropsychologia ，37 ，1155-1163 ， 1999. for cue decoding during spoken language processing. 10) 10) Schirmer ，A. ，Kotz ，S.A. ， et a1.: Sex differentiates the NeuroReport ，9 (5) ，899-903 ， 1998.

-28- Joumal Joumal of Hiroshima Prefectural College of Health Sciences Humanity and Science 5 (1) 21 - 30 2005

26) 26) Imaizumi ，S. ，Homma ，M. ， et al.: Gender differences in 27) Noguchi ，Y. ，Ozawa ，Y. ， et al.: Development of Mind- the the functional organization of the brain for emotional reading Capability from Speech. Japan Joumal of prosody prosody processing ， Speech Prosody 2004 ，605-608 ， Logopedics and Phoniatrics ，45(3) ，269-275 ， 2004. 2004. 2004.

内ノム】ハ ud 広島県立保健福祉大学誌 人間と科学 5( 1) 21-30 2005

話者の真意を理解する脳機構とその発達

本丸問石 緑正2 今泉敏本l 小津由嗣*1 山崎和子*1 治 村中博幸牢2

*1 *1 広島県立保健福祉大学コミュニケーション障害学科 *2 *2 広島県立身体障害者リハビリテーションセンター

2004 年 9 月 10 日受付 2004 年 12 月 13 日受理

抄録

話し言葉の言語的意味と感情的態度を判断する課題に対して24 名から観測された機能的MRI による脳機能画 像と 339 名の課題成績を分析して，音声コミュニケーションにおいて話者の心を理解する脳機能を検討した。言

語的意味が正負で対立するフレーズ(好き VS. 嫌しミ)を 感情的態度が対立するモード(1愛 d情JVS.I 憎しみJ) で 女性話者 1 名が発話した音声を使用した。脳機能画像解析の結果，話者の心の理解には，左右上側頭溝，左右前 頭葉下部，右前頭葉内側部，左右小脳後部などが関与することが示唆された。女性被験者に対して男性で有意に 賦活レベルの高かった右前頭葉内側部は，話者の真意を推測する機能に関与すると考えられた。課題成績は，刺 激の感情属性と言語属性が一致しないフレーズに対して小学生から中学生に渡って加齢とともに有意に上昇した。 女児のほうが男児より発達が早いことも示された。これらの結果は，話者の心を理解する脳機能において，従来 独立と仮定されてきた言語的情報処理と感情的情報処理とが相互に干渉し合うこと この脳機能には性差がある ことを示唆する。

キーワード:音声を介した心の理解脳発達音声コミュニケーション

qtυハ Hv