- Email: [email protected]
Tonotopic maps and short-term plasticity in the human auditory cortex
International Congress Series 1270 (2004) 67 – 73
www.ics-elsevier.com
Tonotopic maps and short-term plasticity in the human auditory cortex Isamu Ozaki * Department of Physical Therapy, Faculty of Health Sciences, Aomori University of Health and Welfare, 58-1 Hamadate, Mase, Aomori 030-8505, Japan
Abstract. We showed tonotopic maps in the human auditory cortex based on the analysis of auditory evoked magnetic fields (AEFs) and rapid changes in tonotopic maps during auditory attention tasks. The location and strength of the N100m dipole for 400- and 4000-Hz tones were successively calculated before and at around the peak latency. In the right hemisphere, the current dipoles for 400 and 4000 Hz moved toward anterolateral direction before the N100m peak, showing parallel arrangement of the isofrequency bands. In the left hemisphere, the movement direction of 400-Hz dipoles was anterolateral, while that of 4000-Hz dipoles was lateral. This difference in the layout of isofrequency bands between right and left auditory cortices reflects distinct functional roles in auditory information processing such as pitch vs. phonetic analysis. In addition, when subjects tried to discriminate the differences in pitch or laterality, N100m amplitude increased. The N100m dipole distance between 400- and 4000-Hz tones was enlarged during pitch discrimination tasks especially in the right auditory cortex but was unchanged during laterality discrimination tasks. These dynamic changes in the N100m dipole presumably reflect a short-term plasticity in the primary auditory cortex. D 2004 Elsevier B.V. All rights reserved. Keywords: Auditory evoked magnetic field; AEF; Isofrequency band; Auditory attention; Tonotopic map; Pitch discrimination; Laterality discrimination
1. Introduction In an earlier study of steady-state auditory evoked magnetic fields (AEFs), Romani et al. [1] proposed that high-frequency tone is represented in the medial portion of Heschl gyrus. However, there remain debates as to tonotopic representation in the human auditory cortex. The N100m dipole for high-frequency tones reportedly is located more medially [2– 6] or more posteriorly [4,7,8] than that for low-frequency tone. Other studies on AEFs claim that there is no tonotopic representation [9– 11]. We suppose that these divergent results on human tonotopy have two sources [12]; one is that the location of the N100m dipole is determined at the peak latency; and the other is that there is a significant inter-individual or inter-hemispheric variety in 3D morphology of the human AI cortex [13]. We have reported * Tel./fax: +81-17-765-2070. E-mail address: [email protected] (I. Ozaki). 0531-5131/ D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.ics.2004.05.083
68
I. Ozaki / International Congress Series 1270 (2004) 67–73
that N100m dipole dynamically moves as a result of sequential activation of the neural columns that form the isofrequency bands [14,15] and that the isofrequency bands extend along the long axis of Heschl gyrus. In the present article, we will show two representative cases in which N100m dipole dynamically moves in a mostly lateral direction or a mostly anterior direction, reflecting the difference in 3D morphology of Heschl gyrus. Also, we will show rapid changes in N100m response during auditory attention. 2. Tonotopic maps as dynamic movement of N100m dipole Here, we will demonstrate two cases from our previous studies [15]. AEFs (band-pass 0.1– 330 Hz) were taken with a Neuromag system (4-D Neuroimaging, Helsinki, Finland), which has 204 planar first-order gradiometers at 102 measurement sites on a helmet-shaped surface, covering the whole scalp. Monaural 400- or 4000-Hz tone pips of 80 dB SPL (sound pressure level) and 50-ms duration with 2 ms rise – fall times were delivered by a plastic tube terminating in a molded ear insert. The stimuli were presented, with an interstimulus interval of 1 s, to the subject’s right or left ear. The N100m current sources at each sampling point during the period between 25 ms before the peak and 5 ms after the peak were calculated from the AEFs recorded from the hemisphere contralateral to the stimulation site using a single equivalent current dipole (ECD) model in a spherical volume conductor. The details about data analysis including 3D MRI study appeared elsewhere [12,14,15]. Fig. 1 shows the case in which N100m dipole moves in a mostly lateral direction. When the AEFs become augmented from 70 to 90 ms, the distribution of the magnetic fields on the lateral view changes dynamically. That is, the distance between the extremas of flux out and flux in is shortening, suggesting that the dipole is approaching the recording device; i.e., the N100m dipole dynamically moves in medio-lateral direction. In Fig. 2, the dipoles for 400- and 4000-Hz tones, calculated sequentially up to the peak latency with continuity of the goodness-of-fit (GOF) value >90%, are superimposed onto the brain MRI of this subject. They are located on the supratemporal plane of the right hemisphere; the dipoles for high-frequency tone are mapped
Fig. 1. (A) Superimposed AEF waveforms of the right hemisphere following left ear 400-Hz-tone stimulation in a 25-year-old man. (B) Magnetic fields at 70, 80 and 90 ms. Note that the distance between the extrema of flux out (thick traces) and flux in (thin traces) is shortening, suggesting that the dipole is approaching the recording device.
I. Ozaki / International Congress Series 1270 (2004) 67–73
69
Fig. 2. Mediolateral movement of the N100m dipole for a 4000-Hz tone (A) and for a 400-Hz tone (B), obtained from the same subject in Fig. 1. Note that the dipoles for high-pitched tone are located more anteriorly.
anteriorly and travel with shorter distance. In the owl monkey AI cortex, the isofrequency bands are roughly parallel to the lateral fissure, the anterolateral border of the AI [16]. It is, therefore, supposed that in humans, the isofrequency bands are roughly parallel to the anterior border of the Heschl gyrus. So, we suggest that in this subject, Heschl gyrus mostly extends in the lateral direction and that the N100m dipoles for high-frequency tone at the peak latency is located more anteriorly. The results of another subject in which N100m dipole moves in a mostly anterior direction are illustrated in Fig. 3. As shown in Fig. 3B, the border between flux out and flux in is displaced in the anterior direction in the rising phase of the N100m response (between 56 and 70 ms poststimulus). The estimated dipoles at 56 and 70 ms with the GOF value >90% are located on the supratemporal plane of the right hemisphere. We, therefore, suppose that in this subject, Heschl gyrus mostly extends in the anterior direction.
Fig. 3. (A) Superimposed AEF waveforms of the right hemisphere following left ear 400-Hz-tone stimulation in a 33-year-old woman. (B) Magnetic fields at 56 and 70 ms and the N100m dipoles from 56 to 70 ms superimposed on to the subject’s MRI. Note that the estimated dipoles (white arrows or circles) move in the anterior direction.
70
I. Ozaki / International Congress Series 1270 (2004) 67–73
Based on analysis of the N100m dipole movement in the rising phase of N100m response recorded from 31 healthy subjects, we obtained the layout of the isofrequency bands for 400 and 4000 Hz as shown in Fig. 4. They are in good agreement with the probability map of the primary auditory cortex based on the morphological analysis of 27 postmortem brains [13]. Firstly, the bands for the left hemisphere are located more posteriorly than for the right hemisphere. Secondly, in the right hemisphere, the isofrequency bands for 400 and 4000 Hz extend in an anterolateral direction. The parallel arrangement of the isofrequency bands for high-pitched and low-pitched tones in the right AI cortex are in line with the distinct functional role of the right hemisphere where the differences in pitch of the tones and music are analyzed [15]. The transverse temporal gyri of Heschl in the postmortem brain have an area of ca. 10 mm width 30 mm length. We suppose that the isofrequency bands extend along the long axis of Heschl gyrus. Since the axis of the frequency is orthogonal to each band for the different frequencies represented, the axis of the frequency for 20 –20,000 Hz tones that are audible for humans presumably occupies up to ca. 10 mm. Since each isofrequency band for the frequency representation is displaced with a function of the logarithm of the frequency [16,17], the isofrequency bands representing the frequencies with a difference of a factor of 10 in human will be ca. 3 mm apart from each other, which is in line with the results of the analysis of AEFs in right auditory cortex (Fig. 4, right panel). Fig. 5 illustrates schematic drawing of tonotopic map when Heschl gyrus extends mostly in anterior direction. In these subjects, the bands for high-pitched tones are located more medially and are shorter than those for low-pitched tones. Therefore, when one determines N100m dipole location at the peak latency, one can obtain the difference in the locations such as Dx and Dy, i.e., a high-pitched tone can be concluded to be located more posteriorly by the difference of Dy. However, this reflects isofrequency band-dependent tonotopy but not classical tonotopy (Dt). If one examines the subjects whose Heschl gyrus mostly extends in lateral direction, tonotopic representation as the locations of the N100m dipoles at the peak latency for different pitches of the tones examined may differ. On the other hand, when
Fig. 4. Isofrequency bands for 400-Hz (closed circle) and 4000-Hz (open square) tones represented as normalized movement of the N100m dipoles on the x – y plane in left and right hemispheres. Arrows indicate the direction of N100m dipole movement from the starting time analyzed to the peak latency. In the right hemisphere, the isofrequency bands for two frequencies are in parallel arrangement and the distance between the two bands is ca. 3 mm (modified from Ref. [15]).
I. Ozaki / International Congress Series 1270 (2004) 67–73
71
Fig. 5. Schematic drawings of the tonotopic maps when a subject’s Heschl gyrus in the right hemisphere extends in a mostly anterior direction (modified from Ref. [12]). In the left panel, the layout of the isofrequency bands for different frequency tones is illustrated in the Heschl gyrus (small arrows, the first transverse temporal sulcus). In the right column, open and closed circles represent the locations of the N100m dipoles at the peak latency for high-pitched and low-pitched tones, respectively. The movements of the N100m dipole along the isofrequency bands are indicated by arrows. On the other hand, the differences in the locations (Dx or Dy) of the N100m dipoles at the peak latency (open and closed circles) for the different tone frequencies indicate ‘‘isofrequency band dependent tonopy’’. ‘‘Classical tonotopy’’ is expressed by Dt that is nearly 3 mm when a difference of the tone frequencies examined is a factor of 10. A, anterior; P, posterior; M, medial; L, lateral.
one determines dynamic movement of N100m dipole, one can obtain Dt, classical tonotopy, as the distance between the bands. To our knowledge, most previous papers on human tonotopy have argued tonotopic representations based on the locations of the N100m dipoles at the peak and neglected a significant inter-subject or inter-hemispheric variety in 3D morphology of Heschl gyrus, resulting in the divergent results on human tonotopy. 3. Rapid changes in tonotopic maps during auditory attention We examined whether tonotopic maps are changed during auditory attention task. Our hypothesis was that since cortical representation of tone is fundamentally distributed but not focal along the primary auditory cortex [17], the activated areas of the AI cortex for the tones with different pitches will be segregated when subjects try to discriminate the difference in pitch of the tones as short-term plastic changes. We also hypothesized that segregation of tonotopic maps for the different pitches will not occur when subjects try to discriminate laterality of the tones but not the differences in pitch. To test the above hypotheses, we examined 23 right-handed, normal subjects [12]; we presented a 400- or a 4000-Hz tone to the subject’s right or left ear with a random sequence. Subjects performed pitch discrimination tasks in which the target was high-pitched (right or left 4000-Hz tone, 10% each) or low-pitched (right or left 400-Hz tone, 10% each) and then laterality discrimination tasks in which the target was right (right 400- or 4000 Hz-tone, 10% each) or left (left 400- or 4000-Hz tone, 10% each). The details of the experimental design and results appear elsewhere [12]. We analyzed N100m responses as control or non-target responses; since in each task, subjects attend four categories of stimuli, non-target responses include effects of attention but not influence of finger extension to the target stimuli. As a result, attention produced an increase of dipole strength in pitch and laterality discriminating condition (Fig. 6) as well as shortening of the peak latency. On the other hand, the Euclidian dipole distance between 400- and 4000-Hz tones was enlarged in pitch discriminating condition especially
72
I. Ozaki / International Congress Series 1270 (2004) 67–73
Fig. 6. Upper panel: the mean dipole strengths of the N100m responses for 400-Hz (closed bars) and 4000-Hz (open bars) tones in control and pitch or laterality discriminating conditions. The mean N100m dipole strength is augmented in pitch and laterality discriminating conditions. Lower panel: the N100m dipole distance between 400and 4000-Hz tones in the three conditions. The mean dipole distance for pitch discriminating condition was larger. Note that enlargement of the dipole strength is marked in the right hemisphere. Results demonstrate mean F S.E.M. N = 18 for the data for the left hemisphere, N = 23 for the right hemisphere (modified from Ref. [12]).
in the right hemisphere, but not in laterality discriminating condition (Fig. 6). These results suggest short-term plastic changes in the human auditory cortex during selective attention, which is in line with experimental data of the owl monkey showing that the accuracy of the performance is correlated with an enlargement of tonotopic map during training to discriminate the small difference of pitch of the tones [18]. Therefore, topographically organized neurons in the primary sensory cortex presumably play an important role in analyzing or storing the features of sensory stimuli. Acknowledgements We thank Drs. C.Y. Jin, Y. Suzuki, M. Baba, M. Matsunaga and I. Hashimoto for their valuable help. This study was supported by a Special Research Project Grant, Aomori University of Health and Welfare. References [1] G.L. Romani, S.J. Williamson, L. Kaufman, Tonotopic organization of the human auditory cortex, Science 216 (1982) 1339 – 1340. [2] C. Pantev, et al., Tonotopic organization of the human auditory cortex revealed by transient auditory evoked magnetic fields, Electroenceph. Clin. Neurophysiol. 69 (1988) 160 – 170. [3] C. Pantev, et al., Specific tonotopic organizations of different areas of the human auditory cortex revealed by simultaneous magnetic and electric recordings, Electroenceph. Clin. Neurophysiol. 94 (1995) 26 – 40. [4] C. Pantev, et al., Study of the human auditory cortices using a whole-head magnetometer: left vs. right hemisphere and ipsilateral vs. contralateral stimulation, Audiol. Neuro-otol. 3 (1998) 183 – 190. [5] M. Huotilainen, et al., Sustained fields of tones and glides reflect tonotopy of the auditory cortex, NeuroReport 6 (1995) 841 – 844. [6] D.C. Rojas, et al., Alterations in tonotopy and auditory cerebral asymmetry in schizophrenia, Biol. Psychiatry 52 (2002) 32 – 39. [7] S. Arlinger, et al., Cortical magnetic fields evoked by frequency glides of a continuous tone, Electroenceph. Clin. Neurophysiol. 54 (1982) 642 – 653. [8] T. Rosburg, et al., Tonotopy of the auditory-evoked field component N100m in patients with schizophrenia, J. Psychophysiol. 14 (2000) 131 – 141. [9] T.P. Roberts, D. Poeppel, Latency of auditory evoked M100 as a function of tone frequency, NeuroReport 7 (1996) 1138 – 1140.
I. Ozaki / International Congress Series 1270 (2004) 67–73
73
[10] B. Lu¨tkenho¨ner, Single-dipole analyses of the N100m are not suitable for characterizing the cortical representation of pitch, Audiol. Neuro-otol. 8 (2003) 222 – 233. [11] B. Lu¨tkenho¨ner, K. Krumbholz, A. Seither-Preisler, Studies of tonotopy based on wave N100 of the auditory evoked field are problematic, NeuroImage 19 (2003) 935 – 949. [12] I. Ozaki, et al., Rapid change of tonotopic maps in the human auditory cortex during pitch discrimination. Clin. Neurophysiol. 115 (2004) 1592 – 1604. [13] J. Rademacher, et al., Probabilistic mapping and volume measurement of human primary auditory cortex, NeuroImage 13 (2001) 669 – 683. [14] I. Ozaki, et al., Dynamic anterolateral movement of N100m dipoles in evoked magnetic field reflects activation of isofrequency bands through horizontal fibers in human auditory cortex, Neurosci. Lett. 329 (2002) 222 – 226. [15] I. Ozaki, et al., Dynamic movement of N100m dipoles in evoked magnetic field reflects sequential activation of isofrequency bands in human auditory cortex, Clin. Neurophysiol. 114 (2003) 1681 – 1688. [16] G.H. Recanzone, et al., Functional organization of spectral receptive fields in the primary auditory cortex of the owl monkey, J. Comp. Neurol. 415 (1999) 460 – 481. [17] C.E. Schreiner, Spatial distribution of responses to simple and complex sounds in the primary auditory cortex, Audiol. Neuro-otol. 3 (1998) 104 – 122. [18] G.H. Recanzone, C.E. Schreiner, M.M. Merzenich, Plasticity in the frequency representation of primary auditory cortex following discrimination training in adult owl monkey, J. Neurosci. 13 (1993) 87 – 103.
www.ics-elsevier.com
Tonotopic maps and short-term plasticity in the human auditory cortex Isamu Ozaki * Department of Physical Therapy, Faculty of Health Sciences, Aomori University of Health and Welfare, 58-1 Hamadate, Mase, Aomori 030-8505, Japan
Abstract. We showed tonotopic maps in the human auditory cortex based on the analysis of auditory evoked magnetic fields (AEFs) and rapid changes in tonotopic maps during auditory attention tasks. The location and strength of the N100m dipole for 400- and 4000-Hz tones were successively calculated before and at around the peak latency. In the right hemisphere, the current dipoles for 400 and 4000 Hz moved toward anterolateral direction before the N100m peak, showing parallel arrangement of the isofrequency bands. In the left hemisphere, the movement direction of 400-Hz dipoles was anterolateral, while that of 4000-Hz dipoles was lateral. This difference in the layout of isofrequency bands between right and left auditory cortices reflects distinct functional roles in auditory information processing such as pitch vs. phonetic analysis. In addition, when subjects tried to discriminate the differences in pitch or laterality, N100m amplitude increased. The N100m dipole distance between 400- and 4000-Hz tones was enlarged during pitch discrimination tasks especially in the right auditory cortex but was unchanged during laterality discrimination tasks. These dynamic changes in the N100m dipole presumably reflect a short-term plasticity in the primary auditory cortex. D 2004 Elsevier B.V. All rights reserved. Keywords: Auditory evoked magnetic field; AEF; Isofrequency band; Auditory attention; Tonotopic map; Pitch discrimination; Laterality discrimination
1. Introduction In an earlier study of steady-state auditory evoked magnetic fields (AEFs), Romani et al. [1] proposed that high-frequency tone is represented in the medial portion of Heschl gyrus. However, there remain debates as to tonotopic representation in the human auditory cortex. The N100m dipole for high-frequency tones reportedly is located more medially [2– 6] or more posteriorly [4,7,8] than that for low-frequency tone. Other studies on AEFs claim that there is no tonotopic representation [9– 11]. We suppose that these divergent results on human tonotopy have two sources [12]; one is that the location of the N100m dipole is determined at the peak latency; and the other is that there is a significant inter-individual or inter-hemispheric variety in 3D morphology of the human AI cortex [13]. We have reported * Tel./fax: +81-17-765-2070. E-mail address: [email protected] (I. Ozaki). 0531-5131/ D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.ics.2004.05.083
68
I. Ozaki / International Congress Series 1270 (2004) 67–73
that N100m dipole dynamically moves as a result of sequential activation of the neural columns that form the isofrequency bands [14,15] and that the isofrequency bands extend along the long axis of Heschl gyrus. In the present article, we will show two representative cases in which N100m dipole dynamically moves in a mostly lateral direction or a mostly anterior direction, reflecting the difference in 3D morphology of Heschl gyrus. Also, we will show rapid changes in N100m response during auditory attention. 2. Tonotopic maps as dynamic movement of N100m dipole Here, we will demonstrate two cases from our previous studies [15]. AEFs (band-pass 0.1– 330 Hz) were taken with a Neuromag system (4-D Neuroimaging, Helsinki, Finland), which has 204 planar first-order gradiometers at 102 measurement sites on a helmet-shaped surface, covering the whole scalp. Monaural 400- or 4000-Hz tone pips of 80 dB SPL (sound pressure level) and 50-ms duration with 2 ms rise – fall times were delivered by a plastic tube terminating in a molded ear insert. The stimuli were presented, with an interstimulus interval of 1 s, to the subject’s right or left ear. The N100m current sources at each sampling point during the period between 25 ms before the peak and 5 ms after the peak were calculated from the AEFs recorded from the hemisphere contralateral to the stimulation site using a single equivalent current dipole (ECD) model in a spherical volume conductor. The details about data analysis including 3D MRI study appeared elsewhere [12,14,15]. Fig. 1 shows the case in which N100m dipole moves in a mostly lateral direction. When the AEFs become augmented from 70 to 90 ms, the distribution of the magnetic fields on the lateral view changes dynamically. That is, the distance between the extremas of flux out and flux in is shortening, suggesting that the dipole is approaching the recording device; i.e., the N100m dipole dynamically moves in medio-lateral direction. In Fig. 2, the dipoles for 400- and 4000-Hz tones, calculated sequentially up to the peak latency with continuity of the goodness-of-fit (GOF) value >90%, are superimposed onto the brain MRI of this subject. They are located on the supratemporal plane of the right hemisphere; the dipoles for high-frequency tone are mapped
Fig. 1. (A) Superimposed AEF waveforms of the right hemisphere following left ear 400-Hz-tone stimulation in a 25-year-old man. (B) Magnetic fields at 70, 80 and 90 ms. Note that the distance between the extrema of flux out (thick traces) and flux in (thin traces) is shortening, suggesting that the dipole is approaching the recording device.
I. Ozaki / International Congress Series 1270 (2004) 67–73
69
Fig. 2. Mediolateral movement of the N100m dipole for a 4000-Hz tone (A) and for a 400-Hz tone (B), obtained from the same subject in Fig. 1. Note that the dipoles for high-pitched tone are located more anteriorly.
anteriorly and travel with shorter distance. In the owl monkey AI cortex, the isofrequency bands are roughly parallel to the lateral fissure, the anterolateral border of the AI [16]. It is, therefore, supposed that in humans, the isofrequency bands are roughly parallel to the anterior border of the Heschl gyrus. So, we suggest that in this subject, Heschl gyrus mostly extends in the lateral direction and that the N100m dipoles for high-frequency tone at the peak latency is located more anteriorly. The results of another subject in which N100m dipole moves in a mostly anterior direction are illustrated in Fig. 3. As shown in Fig. 3B, the border between flux out and flux in is displaced in the anterior direction in the rising phase of the N100m response (between 56 and 70 ms poststimulus). The estimated dipoles at 56 and 70 ms with the GOF value >90% are located on the supratemporal plane of the right hemisphere. We, therefore, suppose that in this subject, Heschl gyrus mostly extends in the anterior direction.
Fig. 3. (A) Superimposed AEF waveforms of the right hemisphere following left ear 400-Hz-tone stimulation in a 33-year-old woman. (B) Magnetic fields at 56 and 70 ms and the N100m dipoles from 56 to 70 ms superimposed on to the subject’s MRI. Note that the estimated dipoles (white arrows or circles) move in the anterior direction.
70
I. Ozaki / International Congress Series 1270 (2004) 67–73
Based on analysis of the N100m dipole movement in the rising phase of N100m response recorded from 31 healthy subjects, we obtained the layout of the isofrequency bands for 400 and 4000 Hz as shown in Fig. 4. They are in good agreement with the probability map of the primary auditory cortex based on the morphological analysis of 27 postmortem brains [13]. Firstly, the bands for the left hemisphere are located more posteriorly than for the right hemisphere. Secondly, in the right hemisphere, the isofrequency bands for 400 and 4000 Hz extend in an anterolateral direction. The parallel arrangement of the isofrequency bands for high-pitched and low-pitched tones in the right AI cortex are in line with the distinct functional role of the right hemisphere where the differences in pitch of the tones and music are analyzed [15]. The transverse temporal gyri of Heschl in the postmortem brain have an area of ca. 10 mm width 30 mm length. We suppose that the isofrequency bands extend along the long axis of Heschl gyrus. Since the axis of the frequency is orthogonal to each band for the different frequencies represented, the axis of the frequency for 20 –20,000 Hz tones that are audible for humans presumably occupies up to ca. 10 mm. Since each isofrequency band for the frequency representation is displaced with a function of the logarithm of the frequency [16,17], the isofrequency bands representing the frequencies with a difference of a factor of 10 in human will be ca. 3 mm apart from each other, which is in line with the results of the analysis of AEFs in right auditory cortex (Fig. 4, right panel). Fig. 5 illustrates schematic drawing of tonotopic map when Heschl gyrus extends mostly in anterior direction. In these subjects, the bands for high-pitched tones are located more medially and are shorter than those for low-pitched tones. Therefore, when one determines N100m dipole location at the peak latency, one can obtain the difference in the locations such as Dx and Dy, i.e., a high-pitched tone can be concluded to be located more posteriorly by the difference of Dy. However, this reflects isofrequency band-dependent tonotopy but not classical tonotopy (Dt). If one examines the subjects whose Heschl gyrus mostly extends in lateral direction, tonotopic representation as the locations of the N100m dipoles at the peak latency for different pitches of the tones examined may differ. On the other hand, when
Fig. 4. Isofrequency bands for 400-Hz (closed circle) and 4000-Hz (open square) tones represented as normalized movement of the N100m dipoles on the x – y plane in left and right hemispheres. Arrows indicate the direction of N100m dipole movement from the starting time analyzed to the peak latency. In the right hemisphere, the isofrequency bands for two frequencies are in parallel arrangement and the distance between the two bands is ca. 3 mm (modified from Ref. [15]).
I. Ozaki / International Congress Series 1270 (2004) 67–73
71
Fig. 5. Schematic drawings of the tonotopic maps when a subject’s Heschl gyrus in the right hemisphere extends in a mostly anterior direction (modified from Ref. [12]). In the left panel, the layout of the isofrequency bands for different frequency tones is illustrated in the Heschl gyrus (small arrows, the first transverse temporal sulcus). In the right column, open and closed circles represent the locations of the N100m dipoles at the peak latency for high-pitched and low-pitched tones, respectively. The movements of the N100m dipole along the isofrequency bands are indicated by arrows. On the other hand, the differences in the locations (Dx or Dy) of the N100m dipoles at the peak latency (open and closed circles) for the different tone frequencies indicate ‘‘isofrequency band dependent tonopy’’. ‘‘Classical tonotopy’’ is expressed by Dt that is nearly 3 mm when a difference of the tone frequencies examined is a factor of 10. A, anterior; P, posterior; M, medial; L, lateral.
one determines dynamic movement of N100m dipole, one can obtain Dt, classical tonotopy, as the distance between the bands. To our knowledge, most previous papers on human tonotopy have argued tonotopic representations based on the locations of the N100m dipoles at the peak and neglected a significant inter-subject or inter-hemispheric variety in 3D morphology of Heschl gyrus, resulting in the divergent results on human tonotopy. 3. Rapid changes in tonotopic maps during auditory attention We examined whether tonotopic maps are changed during auditory attention task. Our hypothesis was that since cortical representation of tone is fundamentally distributed but not focal along the primary auditory cortex [17], the activated areas of the AI cortex for the tones with different pitches will be segregated when subjects try to discriminate the difference in pitch of the tones as short-term plastic changes. We also hypothesized that segregation of tonotopic maps for the different pitches will not occur when subjects try to discriminate laterality of the tones but not the differences in pitch. To test the above hypotheses, we examined 23 right-handed, normal subjects [12]; we presented a 400- or a 4000-Hz tone to the subject’s right or left ear with a random sequence. Subjects performed pitch discrimination tasks in which the target was high-pitched (right or left 4000-Hz tone, 10% each) or low-pitched (right or left 400-Hz tone, 10% each) and then laterality discrimination tasks in which the target was right (right 400- or 4000 Hz-tone, 10% each) or left (left 400- or 4000-Hz tone, 10% each). The details of the experimental design and results appear elsewhere [12]. We analyzed N100m responses as control or non-target responses; since in each task, subjects attend four categories of stimuli, non-target responses include effects of attention but not influence of finger extension to the target stimuli. As a result, attention produced an increase of dipole strength in pitch and laterality discriminating condition (Fig. 6) as well as shortening of the peak latency. On the other hand, the Euclidian dipole distance between 400- and 4000-Hz tones was enlarged in pitch discriminating condition especially
72
I. Ozaki / International Congress Series 1270 (2004) 67–73
Fig. 6. Upper panel: the mean dipole strengths of the N100m responses for 400-Hz (closed bars) and 4000-Hz (open bars) tones in control and pitch or laterality discriminating conditions. The mean N100m dipole strength is augmented in pitch and laterality discriminating conditions. Lower panel: the N100m dipole distance between 400and 4000-Hz tones in the three conditions. The mean dipole distance for pitch discriminating condition was larger. Note that enlargement of the dipole strength is marked in the right hemisphere. Results demonstrate mean F S.E.M. N = 18 for the data for the left hemisphere, N = 23 for the right hemisphere (modified from Ref. [12]).
in the right hemisphere, but not in laterality discriminating condition (Fig. 6). These results suggest short-term plastic changes in the human auditory cortex during selective attention, which is in line with experimental data of the owl monkey showing that the accuracy of the performance is correlated with an enlargement of tonotopic map during training to discriminate the small difference of pitch of the tones [18]. Therefore, topographically organized neurons in the primary sensory cortex presumably play an important role in analyzing or storing the features of sensory stimuli. Acknowledgements We thank Drs. C.Y. Jin, Y. Suzuki, M. Baba, M. Matsunaga and I. Hashimoto for their valuable help. This study was supported by a Special Research Project Grant, Aomori University of Health and Welfare. References [1] G.L. Romani, S.J. Williamson, L. Kaufman, Tonotopic organization of the human auditory cortex, Science 216 (1982) 1339 – 1340. [2] C. Pantev, et al., Tonotopic organization of the human auditory cortex revealed by transient auditory evoked magnetic fields, Electroenceph. Clin. Neurophysiol. 69 (1988) 160 – 170. [3] C. Pantev, et al., Specific tonotopic organizations of different areas of the human auditory cortex revealed by simultaneous magnetic and electric recordings, Electroenceph. Clin. Neurophysiol. 94 (1995) 26 – 40. [4] C. Pantev, et al., Study of the human auditory cortices using a whole-head magnetometer: left vs. right hemisphere and ipsilateral vs. contralateral stimulation, Audiol. Neuro-otol. 3 (1998) 183 – 190. [5] M. Huotilainen, et al., Sustained fields of tones and glides reflect tonotopy of the auditory cortex, NeuroReport 6 (1995) 841 – 844. [6] D.C. Rojas, et al., Alterations in tonotopy and auditory cerebral asymmetry in schizophrenia, Biol. Psychiatry 52 (2002) 32 – 39. [7] S. Arlinger, et al., Cortical magnetic fields evoked by frequency glides of a continuous tone, Electroenceph. Clin. Neurophysiol. 54 (1982) 642 – 653. [8] T. Rosburg, et al., Tonotopy of the auditory-evoked field component N100m in patients with schizophrenia, J. Psychophysiol. 14 (2000) 131 – 141. [9] T.P. Roberts, D. Poeppel, Latency of auditory evoked M100 as a function of tone frequency, NeuroReport 7 (1996) 1138 – 1140.
I. Ozaki / International Congress Series 1270 (2004) 67–73
73
[10] B. Lu¨tkenho¨ner, Single-dipole analyses of the N100m are not suitable for characterizing the cortical representation of pitch, Audiol. Neuro-otol. 8 (2003) 222 – 233. [11] B. Lu¨tkenho¨ner, K. Krumbholz, A. Seither-Preisler, Studies of tonotopy based on wave N100 of the auditory evoked field are problematic, NeuroImage 19 (2003) 935 – 949. [12] I. Ozaki, et al., Rapid change of tonotopic maps in the human auditory cortex during pitch discrimination. Clin. Neurophysiol. 115 (2004) 1592 – 1604. [13] J. Rademacher, et al., Probabilistic mapping and volume measurement of human primary auditory cortex, NeuroImage 13 (2001) 669 – 683. [14] I. Ozaki, et al., Dynamic anterolateral movement of N100m dipoles in evoked magnetic field reflects activation of isofrequency bands through horizontal fibers in human auditory cortex, Neurosci. Lett. 329 (2002) 222 – 226. [15] I. Ozaki, et al., Dynamic movement of N100m dipoles in evoked magnetic field reflects sequential activation of isofrequency bands in human auditory cortex, Clin. Neurophysiol. 114 (2003) 1681 – 1688. [16] G.H. Recanzone, et al., Functional organization of spectral receptive fields in the primary auditory cortex of the owl monkey, J. Comp. Neurol. 415 (1999) 460 – 481. [17] C.E. Schreiner, Spatial distribution of responses to simple and complex sounds in the primary auditory cortex, Audiol. Neuro-otol. 3 (1998) 104 – 122. [18] G.H. Recanzone, C.E. Schreiner, M.M. Merzenich, Plasticity in the frequency representation of primary auditory cortex following discrimination training in adult owl monkey, J. Neurosci. 13 (1993) 87 – 103.