Ackermann et al. have proposed a two-stage neural control model underlying phylogenetic and ontogenetic evolutions of spoken language. Neural machinery at one stage depends upon the development of monosynaptic projections from the motor cortex to cranial nerve nuclei in the brainstem and the other one involves functions of the cortico-basal ganglia circuits. We appreciate this proposal because we have been interested in the contribution of the cortico-basal ganglia circuits to language and associated abilities in humans. Here we want to extend the authors' view, by arguing for potential roles of the cortico-basal ganglia circuits in various aspects of spoken language in adults.
Accumulating evidence indicates that the basal ganglia participate in speech control in humans. However, the roles of the basal ganglia in language control are still unclear. The functionality of the basal ganglia for spoken language perhaps extends beyond the modulation of laryngeal and orofacial movements. We previously showed basal ganglia activity during a cognitive task involving verbal motor imagery, or “inner speech,” in healthy adults (Hanakawa et al. Reference Hanakawa, Honda, Sawamoto, Okada, Yonekura, Fukuyama and Shibasaki2002). This basal ganglia activity was accompanied by activity in other speech-related brain regions such as supplementary motor area and frontal opercular regions. Moreover, we reported that performance of this verbal imagery task was impaired in patients with basal ganglia dysfunctions (Parkinson's disease) in comparison with matched control participants (Sawamoto et al. Reference Sawamoto, Honda, Hanakawa, Fukuyama and Shibasaki2002). A neuroimaging experiment supported that the impaired performance of the verbal imagery task in Parkinson's disease was associated with dysfunctions of the basal ganglia, the caudate nucleus in particular (Sawamoto et al. Reference Sawamoto, Honda, Hanakawa, Aso, Inoue, Toyoda, Ishizu, Fukuyama and Shibasaki2007). Considering that motor imagery is closely related to motor planning (Hanakawa et al. Reference Hanakawa, Dimyan and Hallett2008), the contribution of the basal ganglia to spoken language likely involves a planning stage of speech.
Of even more importance is to understand the contribution of the basal ganglia to learning of spoken language. Ackermann et al. propose a fundamentally different role of the basal ganglia at ontogenetic stages: acquisition of articulatory motor patterns during childhood versus emotive-prosodic modulation of verbal utterances during adulthood. We want to modify and extend this view, especially with regard to the contrast between childhood and adulthood stages. The neural underpinnings for the native language development are difficult to study experimentally. Therefore, we want to argue for the role of the basal ganglia in speech acquisition in adults, taking the case of second language (L2) learning as an example.
We recently conducted a cohort study in which Japanese university students were enrolled in a 16-week e-learning program to develop their English vocabulary (Hosoda et al. Reference Hosoda, Tanaka, Nariai, Honda and Hanakawa2013). Although the training program involved various aspects of vocabulary learning, an emphasis was placed upon the training of pronunciation. The students learned 60 words or idioms in each week. An example sentence for each word and idiom was also presented. The participants were encouraged to dictate each word, idiom, and sentence 10 times in reference to “speech templates” provided by the program. By repeating after the speech templates, the participants were to compare their own utterances and the speech templates, and then try to make corrections to his or her motor programs for pronunciation. Speculatively, this auditory feedback learning should help the trainees achieve adequate spatio-temporal control of laryngeal and orofacial musculature. After 16 weeks, the trainees showed approximately 30% improvement in a test battery of English competence. We performed multidimensional imaging assessment for neuroplastic changes associated with the training. Most notably, probabilistic diffusion tractography demonstrated that connectivity between the inferior frontal gyrus and the caudate nucleus, an input station of the basal ganglia, was enhanced in correlation with the improvement in the trainees' L2 competence. This study has provided the first evidence that the cortico-basal ganglia circuits are involved in language learning in adults. Furthermore, the learning-induced enhancement of the cortico-basal ganglia connectivity was accompanied by enhanced connectivity between the inferior frontal gyrus and superior temporal/supramarginal gyrus (dorsal pathway), but not between the inferior frontal gyrus and middle temporal gyrus (ventral pathway). The dorsal pathway primarily concerns phonological aspects of language control. Hence, the selective involvement of the dorsal pathway indicated that our training program primarily tapped into phonological aspects of L2 vocabulary.
According to the findings in our study (Hosoda et al. Reference Hosoda, Tanaka, Nariai, Honda and Hanakawa2013), we suggest a possibility that the basal ganglia may contribute to learning of spoken language even in adults. Speech is acquired through experiences of adequate auditory inputs, which is evident in children with hearing loss (Tye-Murray et al. Reference Tye-Murray, Spencer and Woodworth1995). In addition, we suspect that reinforcement-type learning (Demirezen Reference Demirezen1988) subserved by functions of the cortico-basal ganglia circuits may underlie experience-based shaping of spoken language. To improve speech control, it is reasonable for both child and adult learners to rely on information about the success or failure of their speech production. We speculate that the feedback information could be self-generated in adult learners who are enrolled in an e-learning program or be given by family and community members as praise or approval to children. Feedback information indicating successful speech production can be utilized as a positive reinforcer to strengthen the neural circuits a trainee had just activated. The striatum that receives both contextual information from the cortex and reward signals from dopaminergic neurons occupies the best position for reinforcement learning or reward-based temporal difference learning (Doya Reference Doya2008). It would be extremely interesting to figure out the learning stage at which genetic predispositions such as FOXP2 play fundamental roles.
Other studies in bilinguals have shown that the caudate nucleus is important for monitoring and controlling of the two languages in use (Crinion et al. Reference Crinion, Turner, Grogan, Hanakawa, Noppeney, Devlin, Aso, Urayama, Fukuyama, Stockton, Usui, Green and Price2006; Hernandez et al. Reference Hernandez, Dapretto, Mazziotta and Bookheimer2001; Hosoda et al. Reference Hosoda, Hanakawa, Nariai, Ohno and Honda2012).
In conclusion, we generally warrant Ackermann et al.'s proposal that the cortico-basal ganglia circuits may play essential roles in evolutions of spoken language. We, however, consider that the cortico-basal ganglia circuits may contribute to various aspects of spoken language including planning, learning, and controlling of speech in both childhood and adulthood.
Ackermann et al. have proposed a two-stage neural control model underlying phylogenetic and ontogenetic evolutions of spoken language. Neural machinery at one stage depends upon the development of monosynaptic projections from the motor cortex to cranial nerve nuclei in the brainstem and the other one involves functions of the cortico-basal ganglia circuits. We appreciate this proposal because we have been interested in the contribution of the cortico-basal ganglia circuits to language and associated abilities in humans. Here we want to extend the authors' view, by arguing for potential roles of the cortico-basal ganglia circuits in various aspects of spoken language in adults.
Accumulating evidence indicates that the basal ganglia participate in speech control in humans. However, the roles of the basal ganglia in language control are still unclear. The functionality of the basal ganglia for spoken language perhaps extends beyond the modulation of laryngeal and orofacial movements. We previously showed basal ganglia activity during a cognitive task involving verbal motor imagery, or “inner speech,” in healthy adults (Hanakawa et al. Reference Hanakawa, Honda, Sawamoto, Okada, Yonekura, Fukuyama and Shibasaki2002). This basal ganglia activity was accompanied by activity in other speech-related brain regions such as supplementary motor area and frontal opercular regions. Moreover, we reported that performance of this verbal imagery task was impaired in patients with basal ganglia dysfunctions (Parkinson's disease) in comparison with matched control participants (Sawamoto et al. Reference Sawamoto, Honda, Hanakawa, Fukuyama and Shibasaki2002). A neuroimaging experiment supported that the impaired performance of the verbal imagery task in Parkinson's disease was associated with dysfunctions of the basal ganglia, the caudate nucleus in particular (Sawamoto et al. Reference Sawamoto, Honda, Hanakawa, Aso, Inoue, Toyoda, Ishizu, Fukuyama and Shibasaki2007). Considering that motor imagery is closely related to motor planning (Hanakawa et al. Reference Hanakawa, Dimyan and Hallett2008), the contribution of the basal ganglia to spoken language likely involves a planning stage of speech.
Of even more importance is to understand the contribution of the basal ganglia to learning of spoken language. Ackermann et al. propose a fundamentally different role of the basal ganglia at ontogenetic stages: acquisition of articulatory motor patterns during childhood versus emotive-prosodic modulation of verbal utterances during adulthood. We want to modify and extend this view, especially with regard to the contrast between childhood and adulthood stages. The neural underpinnings for the native language development are difficult to study experimentally. Therefore, we want to argue for the role of the basal ganglia in speech acquisition in adults, taking the case of second language (L2) learning as an example.
We recently conducted a cohort study in which Japanese university students were enrolled in a 16-week e-learning program to develop their English vocabulary (Hosoda et al. Reference Hosoda, Tanaka, Nariai, Honda and Hanakawa2013). Although the training program involved various aspects of vocabulary learning, an emphasis was placed upon the training of pronunciation. The students learned 60 words or idioms in each week. An example sentence for each word and idiom was also presented. The participants were encouraged to dictate each word, idiom, and sentence 10 times in reference to “speech templates” provided by the program. By repeating after the speech templates, the participants were to compare their own utterances and the speech templates, and then try to make corrections to his or her motor programs for pronunciation. Speculatively, this auditory feedback learning should help the trainees achieve adequate spatio-temporal control of laryngeal and orofacial musculature. After 16 weeks, the trainees showed approximately 30% improvement in a test battery of English competence. We performed multidimensional imaging assessment for neuroplastic changes associated with the training. Most notably, probabilistic diffusion tractography demonstrated that connectivity between the inferior frontal gyrus and the caudate nucleus, an input station of the basal ganglia, was enhanced in correlation with the improvement in the trainees' L2 competence. This study has provided the first evidence that the cortico-basal ganglia circuits are involved in language learning in adults. Furthermore, the learning-induced enhancement of the cortico-basal ganglia connectivity was accompanied by enhanced connectivity between the inferior frontal gyrus and superior temporal/supramarginal gyrus (dorsal pathway), but not between the inferior frontal gyrus and middle temporal gyrus (ventral pathway). The dorsal pathway primarily concerns phonological aspects of language control. Hence, the selective involvement of the dorsal pathway indicated that our training program primarily tapped into phonological aspects of L2 vocabulary.
According to the findings in our study (Hosoda et al. Reference Hosoda, Tanaka, Nariai, Honda and Hanakawa2013), we suggest a possibility that the basal ganglia may contribute to learning of spoken language even in adults. Speech is acquired through experiences of adequate auditory inputs, which is evident in children with hearing loss (Tye-Murray et al. Reference Tye-Murray, Spencer and Woodworth1995). In addition, we suspect that reinforcement-type learning (Demirezen Reference Demirezen1988) subserved by functions of the cortico-basal ganglia circuits may underlie experience-based shaping of spoken language. To improve speech control, it is reasonable for both child and adult learners to rely on information about the success or failure of their speech production. We speculate that the feedback information could be self-generated in adult learners who are enrolled in an e-learning program or be given by family and community members as praise or approval to children. Feedback information indicating successful speech production can be utilized as a positive reinforcer to strengthen the neural circuits a trainee had just activated. The striatum that receives both contextual information from the cortex and reward signals from dopaminergic neurons occupies the best position for reinforcement learning or reward-based temporal difference learning (Doya Reference Doya2008). It would be extremely interesting to figure out the learning stage at which genetic predispositions such as FOXP2 play fundamental roles.
Other studies in bilinguals have shown that the caudate nucleus is important for monitoring and controlling of the two languages in use (Crinion et al. Reference Crinion, Turner, Grogan, Hanakawa, Noppeney, Devlin, Aso, Urayama, Fukuyama, Stockton, Usui, Green and Price2006; Hernandez et al. Reference Hernandez, Dapretto, Mazziotta and Bookheimer2001; Hosoda et al. Reference Hosoda, Hanakawa, Nariai, Ohno and Honda2012).
In conclusion, we generally warrant Ackermann et al.'s proposal that the cortico-basal ganglia circuits may play essential roles in evolutions of spoken language. We, however, consider that the cortico-basal ganglia circuits may contribute to various aspects of spoken language including planning, learning, and controlling of speech in both childhood and adulthood.