Structural Neuroplasticity Effects of Singing in Chronic Aphasia

Aleksi J. Sihvonen; Anni Pitkäniemi; Sini-Tuuli Siponkoski; Linda Kuusela; Noelia Martínez-Molina; Sari Laitinen; Essi-Reetta Särkämö; Johanna Pekkola; Susanna Melkas; Gottfried Schlaug; Viljami Sairanen; Teppo Särkämö

doi:10.1523/ENEURO.0408-23.2024

Article Figures & Data

Figures

Tables
Extended Data

Download figure
Open in new tab
Download powerpoint
Figure 1.
Flowchart.
Download figure
Open in new tab
Download powerpoint
Figure 2.
Lesion overlap map of all patients (N = 28). The n-value represents the number of patients with a lesion to a specific voxel.
Download figure
Open in new tab
Download powerpoint
Figure 3.
Treatment-induced WM neuroplasticity changes. Connectometry results displaying the significant segments of the tracts with longitudinal QA increases significantly associated with singing group versus control group between T1 and T2 (ΔT2–T1; left) and longitudinal QA change correlation with improved naming (right). FDR, false discovery rate; L, left; QA, quantitative anisotropy; R, right.
Download figure
Open in new tab
Download powerpoint
Figure 4.
Treatment-induced GM neuroplasticity changes. Longitudinal GM volume increases (singing group > control group) in T2 > T1 and longitudinal GM volume change correlation with improved naming. Additional exploratory voxel-wise analyses are reported in Extended Data Figure 4-1. Bar plots for mean group GM volume changes are shown: bar, mean; error bar, standard error of mean. BA, Brodmann area; FDR, false discovery rate; vPMC, ventral premotor cortex.

View popup

Table 1.

Demographic and clinical characteristics of the patients

	All	Singing group	Control group	p-value
	N = 28	N = 13	N = 15	p-value
Demographical
Age (years)	64.7 (8.4)	64.1 (8.8)	65.2 (8.3)	0.731 (t)
Sex (male/female)	14/14	7/6	7/8	0.705 (χ²)
Handedness (right/left)	23/5	12/1	11/4	0.191 (χ²)
Education level^{^a}	3.0 (1.3)	3.2 (1.4)	2.9 (1.3)	0.492 (U)
Music background
Choir singing years	3.3 (9.3)	3.8 (12.4)	2.8 (6.0)	0.389 (U)
Singing lessons years	0.6 (3.5)	0.0 (0.0)	1.2 (4.6)	0.352 (U)
Instrument lessons years	0.8 (2.3)	0.2 (0.6)	1.3 (3.0)	0.325 (U)
Clinical
Time from injury (months)	78.1 (82.0)	69.7 (77.9)	85.4 (87.3)	0.982 (U)
Lesion size (cm³)	89.3 (85.3)	119.0 (75.6)	63.5 (87.1)	0.023 (U)
Type of stroke (ischemic/hemorrhagic)	20/8	11/2	9/6	0.150 (χ²)
AQ score (mild or moderate/severe)^{^b}	20/8	8/5	12/3	0.281 (χ²)
BDAE verbal agility score	7.7 (4.8)	6.5 (5.2)	8.8 (4.2)	0.339 (U)
Received rehabilitation (T1–T2)
Speech therapy^{^c}	7.3 (10.8)	7.6 (11.6)	6.9 (10.4)	0.717 (U)
Physical therapy^{^c}	6.9 (12.3)	8.5 (15.7)	5.5 (8.8)	1.000 (U)
Occupational therapy^{^c}	2.2 (5.7)	3.9 (8.0)	0.7 (1.5)	0.525 (U)
Neuropsychological rehabilitation^{^c}	0.3 (1.2)	0.0 (0.0)	0.6 (1.6)	0.555 (U)

Data are mean (SD) unless otherwise reported. Bold values denote statistical significance at p < 0.05. t, t test; χ², chi-square test; U, Mann–Whitney U test; AQ, aphasia quotient; BDAE, Boston Diagnostic Aphasia Examination.
↵a Education level according to the UNESCO International Standard Classification of Education: range 1 (primary education) to 6 (doctoral or equivalent level).
↵b Aphasia severity based on the AQ score: 0–50 = severe, 51–100 = mild/moderate.
↵c Data are mean (SD) in hours between T1 and T2.

View popup

Table 2.

Spoken language and neuroplasticity analysis outcomes

Measure	Group	T1 mean (SD)	T2 mean (SD)	ΔT2 > T1 mean (SD)	Baseline diff. (p_FDR-value)	ΔT2 > T1 (p_FDR-value)
Spoken language outcomes
WAB Naming	Singing	5.0 (3.6)	5.7 (3.8)	0.7 (0.7)	0.585	0.006
WAB Naming	Control	7.4 (3.6)	7.4 (3.5)	0.0 (0.5)	0.585	0.006
WAB Repetition	Singing	5.4 (3.6)	5.7 (3.4)	0.3 (0.7)	0.585	0.277
WAB Repetition	Control	7.6 (3.4)	7.6 (3.3)	0.0 (0.4)	0.585	0.277
Neuroplasticity outcomes
GMV left BA44^{^a}	Singing	747 (568)	807 (586)	61 (108)	0.945	0.008
GMV left BA44^{^a}	Control	854 (387)	713 (427)	−140 (141)	0.945	0.008
GMV right BA44^{^a}	Singing	1,619 (159)	1,628 (278)	9 (99)	0.718	0.587
GMV right BA44^{^a}	Control	1,798 (336)	1,743 (324)	−55 (149)	0.718	0.587
GMV left BA45^{^a}	Singing	387 (304)	414 (306)	28 (43)	0.945	0.024
GMV left BA45^{^a}	Control	409 (240)	336 (263)	−73 (78)	0.945	0.024
GMV right BA45^{^a}	Singing	873 (159)	891 (173)	19 (61)	0.720	0.759
GMV right BA45^{^a}	Control	924 (201)	913 (164)	−11 (110)	0.720	0.759
GMV left vPMC^{^a}	Singing	884 (567)	941 (607)	57 (203)	0.718	0.024
GMV left vPMC^{^a}	Control	1,106 (400)	913 (460)	−195 (151)	0.718	0.024
GMV right vPMC^{^a}	Singing	1,783 (293)	1,775 (309)	−8 (105)	0.833	0.876
GMV right vPMC^{^a}	Control	1,827 (448)	1,810 (370)	−21 (172)	0.833	0.876
GMV left pMTG^{^a}	Singing	1,103 (718)	1,090 (724)	−13 (76)	0.718	0.268
GMV left pMTG^{^a}	Control	1,383 (430)	1,431 (591)	47 (56)	0.718	0.268
GMV right pMTG^{^a}	Singing	2,205 (533)	2,242 (551)	37 (137)	0.718	0.777
GMV right pMTG^{^a}	Control	2,498 (503)	2,513 (477)	14 (74)	0.718	0.777
WM left AF^{^b}	Singing	0.10 (0.04)	0.11 (0.05)	0.01 (0.01)	0.480	0.005
WM left AF^{^b}	Control	0.11 (0.04)	0.11 (0.05)	0.00 (0.02)	0.480	0.005
WM left FAT^{^b}	Singing	0.13 (0.03)	0.14 (0.04)	0.01 (0.02)	0.523	0.004
WM left FAT^{^b}	Control	0.14 (0.05)	0.13 (0.06)	−0.01 (0.02)	0.523	0.004
WM right FAT^{^b}	Singing	0.19 (0.03)	0.20 (0.03)	0.01 (0.01)	0.815	0.004
WM right FAT^{^b}	Control	0.20 (0.03)	0.19 (0.02)	0.00 (0.02)	0.815	0.004
WM left SLF^{^b}	Singing	0.10 (0.05)	0.11 (0.05)	0.01 (0.01)	0.480	0.005
WM left SLF^{^b}	Control	0.11 (0.04)	0.11 (0.05)	0.00 (0.02)	0.480	0.005
WM right SLF^{^b}	Singing	0.21 (0.03)	0.22 (0.03)	0.01 (0.01)	0.480	0.004
WM right SLF^{^b}	Control	0.20 (0.03)	0.20 (0.03)	0.00 (0.01)	0.480	0.004
WM left CS^{^b}	Singing	0.09 (0.04)	0.10 (0.04)	0.01 (0.01)	0.480	0.004
WM left CS^{^b}	Control	0.10 (0.04)	0.10 (0.05)	0.00 (0.01)	0.480	0.004
WM right CS^{^b}	Singing	0.16 (0.04)	0.17 (0.04)	0.01 (0.01)	0.555	0.004
WM right CS^{^b}	Control	0.17 (0.03)	0.17 (0.02)	0.00 (0.01)	0.555	0.004
WM CC^{^b}	Singing	0.14 (0.03)	0.15 (0.03)	0.01 (0.01)	0.480	0.005
WM CC^{^b}	Control	0.15 (0.03)	0.15 (0.03)	−0.01 (0.01)	0.480	0.005

Data are mean (SD) unless otherwise reported. Bold values denote statistical significance at p < 0.05. P-values are FDR corrected.
↵a Volume in mm³.
↵b Normalized QA.

Extended Data

Figures
Tables

Figure 4-1
Voxel-wise analysis of the treatment-induced grey matter neuroplasticity changes. Longitudinal GM volume increases (Singing group>Control Group) in T2>T1 from (left) the whole-brain voxel-wise analysis, (middle) the voxel-wise analysis within the language network derived from meta-analysis (https://neurosynth.org/analyses/terms/language/) and (right) the voxel-wise analysis within the study-specific regions of interest. BA=Brodmann area, FWE=Family-wise error rate, vPMC=ventral premotor cortex. Download Figure 4-1, TIF file.