Absolute Pitch and Planum Temporale - …

1 Absolute Pitch and Planum TemporaleJulian Paul Keenan,* Ven Thangaraj,* Andrea R. Halpern, and Gottfried Schlaug**Department of Neurology, Beth Israel Deaconess Medical Center, 330 Brookline Avenue, Boston, Massachusetts 02215; and Department of Psychology, Bucknell University, Lewisburg, Pennsylvania 17837 Received February 20, 2001: published online October 24, 2001An increased leftward asymmetry of the planumtemporale (PT) in Absolute - Pitch (AP) musicians hasbeen previously reported, with speculation that earlyexposure to music influences the degree of PT asym-metry. To test this hypothesis and to determinewhether a larger left PT or a smaller right PT actuallyaccounts for the increased overall PT asymmetry inAP musicians, anatomical magnetic resonance imageswere taken from a right-handed group of 27 AP musi-cians, 27 nonmusicians, and 22 non-AP musicians. Asignificantly greater leftward PT asymmetry and a sig-nificantly smaller right Absolute PT size for the APmusicians compared to the two control groups wasfound, while the left PT was only marginally larger inthe AP group.

2 The Absolute size of the right PT and notthe left PT was a better predictor of music group mem-bership, possibly indicating pruning of the right PTrather than expansion of the left underlying the in-creased PT asymmetry in AP musicians. Althoughearly exposure to music may be a prerequisite for ac-quiring AP, the increased PT asymmetry in AP musi-cians may be determinedin utero,implicating possiblegenetic influences on PT asymmetry. This may explainwhy the increased PT asymmetry of AP musicians wasnot seen in the group of early beginning non-APmusicians. 2001 Academic PressKey words:music; Absolute Pitch ; morphometry; pla-num Temporale ; Pitch (AP) is the ability to identify the pitchof any tone in the absence of a musical context orreference tone (Takeuchi and Hulse, 1993). The mech-anism and neural basis underlying this ability remainunclear. There is evidence that AP is highly correlatedwith early exposure to music (Krumhansl, 1991;Miyazaki, 1988, 1989) and it has been suggested thatAP manifests itself in a critical period of brain devel-opment (Sergeant, 1969; Schlaug, 1995a)One region that is of interest in regard to AP is theplanum Temporale (PT), a region historically associ-ated with language and auditory processing.

3 The PT isfound to be asymmetric in normal right-handed sam-ples, with a greater leftward bias ( , ). ThePT has been frequently used as a marker of hemi-spheric dominance comparing both normal and clinicalpopulations. Although there are several studies thathave found differences in PT asymmetries dependingon handedness (Foundaset al.,1994; Shapleskeet al.,1999; Steinmetzet al.,1991; Tzourioet al.,1998), thereis still an ongoing dispute in the literature overwhether there is a gender effect in PT asymmetry(Foundaset al.,1999; Ideet al.,1996; Ja ncke andSteinmetz, 1993; Kulynychet al.,1994). Similarly, it isunresolved if PT differences exist between normal in-dividuals and schizophrenics (Bartaet al.,1997; Klein-schmidtet al.,1994; Shapleskeet al.,1999) or betweennormal individuals and dyslexics (Leonardet al.,1993;Shapleskeet al.,1999). Common to each of these stud-ies are the behavioral implications of hemisphericasymmetries of brain regions implicated in languageand, more general, in auditory regard to AP, it has been suggested that there aredifferences in PT asymmetry between AP musiciansand matched non-AP controls.

4 Schlaug and colleagues(1995b) found a greater leftward PT asymmetry be-tween AP musicians and non-AP musicians. Zatorreand colleagues (1998) found a greater leftward PTasymmetry when AP musicians were compared to alarge control sample, although no significant differ-ences in the asymmetry score were found between asmall sample of AP and a small sample of non-APmusicians. These studies demonstrating a possible in-creased leftward asymmetry in AP musicians were in-triguing, considering the usual association of musicalprocessing with right-hemisphere activity changes innonmusicians or individuals unselected for musicalbackground (Zatorreet al.,1992, 1994). However, sev-eral studies questioned whether this association wastrue in musicians (Beveret al.,1974; Mazziottaet al.,1982; Bessonet al.,1994).These findings have raised a series of importantquestions. First, the covariation between a particularmusical ability ( , AP) and the PT has been an un-suspected finding that needs to be replicated in a largersample to address its consistency and further explorethe nature of asymmetric PT differences.

5 One impor-NeuroImage14,1402 1408 (2001) , available online at on14021053-8119/01 $ 2001 by Academic PressAll rights of reproduction in any form , yet not fully examined, possibility is whether theincreased PT asymmetry in AP musicians is due to anincrease in the left PT, a decrease in the right PT, or aredistribution and reallocation of resources betweenthe hemispheres. Galaburda and colleagues speculatedthat the right PT was critical in determining observedPT asymmetries (Galaburdaet al.,1987; Galaburda,1994). In terms of AP, Zatorre and colleagues (1998)found a significant difference in the left PT volumebetween a small group of AP musicians compared tonon-AP musicians unselected for musical skill, but nosignificant difference was found for the right PT inthose group comparisons. However, our previous study(Schlauget al.,1995b) in a much smaller sample thanreported in this study demonstrated a trend for asmaller right PT and a larger left PT, relative to bothnon-AP musicians and nonmusicians.

6 This raises thepossibility that hemispheric specialization might in-volve a redistribution of resources across the hemi-spheres and not just an expansion of one , and equally important, is an understandingof the determinants of an increased leftward PT asym-metry in AP musicians. It is unknown whether theincreased leftward asymmetry is due to early exposureto music and early musical training or is determinedprenatally. It has been suggested that early musicalexposure during critical developmental periods maylead to morphometric differences in motor-relatedbrain structures (Schlauget al.,1995a; Amuntset al.,1997). It is unknown whether early musical exposureor the AP ability per se led to morphometric differ-ences. In order to determine whether early exposure tomusic might have an effect on PT asymmetry or PTsizes, we have tested two large control groups, onegroup of matched (handedness, age, and gender) non-musicians and one group of early beginning musicianswithout Absolute Pitch but with a similar early com-mencement of musical examined a sample of 27 right-handed (12 M/15F; years, ) classically trainedmusicians (mainly keyboard players; some also playedstring instruments in addition) with self-reported ab-solute Pitch ability (AP mus.)

7 The first control groupconsisted of 27 right-handed nonmusicians (12 M/15 F; years, ). Most of the nonmu-sician controls (Nonmus.) were students at local col-leges or universities or professionals in the medicalfield. Nonmusicians were defined as having had noformal training in music or never having played amusical instrument. All were newly recruited from theBoston second control group consisted of 22 right-handed (13 M/9 F; years, )non- Absolute - Pitch musicians (Non-AP mus.). The non- Absolute - Pitch musicians were taken from a databankof over 70 musicians collected over the past 5 selection criteria were non- Absolute - Pitch musi-cian, right-handed, and commencement of musicaltraining before the age of 7. The matching for gender inthis group was not optimal, but we have yet to find agender effect in our group with regard to PT asymme-try (Ja ncke and Steinmetz, 1993), and a recent meta-analysis (Shapleske, 1999) did not find a significantdifference in the Absolute size nor in the differencesbetween left and right PT size between men Absolute Pitch and non- Absolute - Pitch musiciangroups were matched with regard to commencement ofmusical training (AP , , range3 7 years of age; Non-AP , ,range 3 7 years of age;t(47) , ).

8 Theywere also matched in terms of the number of years ofmusical instrument experience (AP , ; Non-AP , ;t(47) , ).Right-handedness was confirmed in all subjects withstandardized tests (Annett, 1992; Ja nckeet al.,1997).All of the subjects gave written informed consent tothis study according to institutional Pitch TestsAbsolute Pitch ability was assessed using two estab-lished tests (Ward and Burns, 1982; Zatorre and Beck-ett, 1989). In both tests, computer-generated sine wavetones with a duration of 500 ms, and an attack anddecay rate of 30 ms, were presented binaurally. Alltones corresponded to fundamental frequencies. Sub-jects were made familiar with the computer-generatedsine wave tones by listening to a sample sequenceconsisting of 10 tones. The first AP test consisted of 13tones taken from a chromatic scale (F#3 to F#4); eachtone was repeated four times in random order with a4-s interstimulus interval. Subjects were required towrite down the tone name in the 4-s interstimulusinterval.

9 After completing the first run through thetotal of 52 tones, subjects had to listen to the tones asecond time and were allowed to make corrections ifneeded. No reference tone was provided at any pointduring the tests and subjects were not informed abouttheir second AP test consisted of a total of 48 sinewave tones taken from 4 1/2 octaves (F2 to G5). First,a set of 12 tones was generated, separated by a musicalfourth (F2, Bb2, Eb3, etc.), so that there was no dupli-cation of chroma across octaves, but large degrees ofdifferences in tone height. The resulting 12 tones werepresented four times in random order for a total of 481403 Absolute Pitch AND Planum Temporale tones. Interstimulus interval was similar to the firsttest. Again, all subjects listened to the whole sequenceof tones two times. In agreement with the literature(Miyazaki, 1988; Wardet al.,1996; Zatorre and Beck-ett, 1989), we regarded a response within 1/2 tonedifference of the presented tone as a correct order to be regarded as an Absolute Pitch possessor,subjects had to have more than 90% correct of theirresponses on both tests.

10 All of the musicians whoclaimed to have Absolute Pitch were confirmed by thesetests ( , all AP musicians in this study scored 90% orhigher correct responses on both of these tests). Themean correct responses were (SD ) for the set of52 tones and (SD ) for the set of 48 tones. Noneof the non-AP musicians and none of the nonmusicianswere able to perform these two tests, since they werenot able to identify any of the tones without a referencetone which we did not provide. All non-AP musicianswere allowed to listen to the tones, but they typicallygave up after the first few tones, since they were notable to identify any of these tones Data Acquisition and Morphometric AnalysesMagnetic resonance imaging was performed employ-ing a T1-weighted magnetization prepared, rapid ac-quisition gradient-echo (MPRAGE) pulse sequence ona Siemens Vision MR system with echo-planarimaging capability (Siemens Medical Systems, Erlan-gen, Germany). A three-plane scout was acquired priorto the MPRAGE to ensure that the interhemisphericplane was aligned with the sagittal plane of case this was not achieved, the subject was reposi-tioned.