Josephs, O. | Search bioRxiv

bioRxiv2018

Activity-is better than connectivity-neurofeedback training in Huntington’s disease

Non-invasive methods, such as neurofeedback training (NFT), could support cognitive symptom management in Huntingtons disease (HD) by targeting brain regions whose function is impaired. The aim of our single-blind, sham-controlled study was to collect rigorous evidence regarding the feasibility of NFT in HD by examining two different methods, activity and connectivity real-time fMRI NFT. Thirty-two HD gene-carriers completed 16 runs of NFT training, using an optimized real-time fMRI protocol. Participants were randomized into four groups, two treatment groups, one receiving neurofeedback derived from the activity of the Supplementary Motor Area (SMA), and another receiving neurofeedback based on the correlation of SMA and left striatum activity (connectivity NFT), and two sham control groups, matched to each of the treatment groups. We examined differences between the groups during NFT training sessions and after training at follow-up sessions. Transfer of training was measured by measuring the participants ability to upregulate NFT target levels without feedback (near transfer), as well as by examining change in objective, a-priori defined, behavioural measures of cognitive and psychomotor function (far transfer) before and at 2 months after training. We found that the treatment group had significantly higher NFT target levels during the training sessions compared to the control group. However, we did not find robust evidence of better transfer in the treatment group compared to controls, or a difference between the two NFT methods. We also did not find evidence in support of a relationship between change in cognitive and psychomotor function and NFT learning success. We conclude that although there is evidence that NFT can be used to guide participants to regulate the activity and connectivity of specific regions in the brain, evidence regarding transfer of learning and clinical benefit was not robust. Although the intervention is non-invasive, given the costs and absence of reliable evidence of clinical benefit, we cannot recommend real-time fMRI NFT as a potential intervention in HD.

neuroscience↗

bioRxiv2017

Distinct neural systems for the production of communicative vocal behaviors

Vocalizations are the production of sounds by the coordinated activity of up to eighty respiratory and laryngeal muscles. Whilst voiced acts, modified by the upper vocal tract (tongue, jaw, lip and palate) are central to the production of human speech, they are also central to the production of emotional vocalizations such as sounds of disgust, anger, laughter and crying. Evidence suggests that the speech and emotional vocalizations may comprise distinct classes of vocal movements: patients with speech production deficits are often able to produce emotional vocalizations. In addition to this, the ontogeny of the two articulations is largely distinct, with some culturally universal emotional vocalizations emerging soon after birth and human speech being a culture specific, highly learnt motor act, which must develop to some degree before a critical period in development. Here we investigate the motor pathways underlying these two classes of vocal movements for the first time by directly comparing BOLD responses during production of speech and non-speech vocal movements. Using functional magnetic resonance imaging, we report distinct patterns of activity in both subcortical and cortical regions (putamen and bilateral inferior frontal and parietal cortices) during the production of emotional vocalizations compared to speech production. In contrast we show that responses in primary sensorimotor regions do not differ during the production of speech and emotional vocalizations, suggesting partially overlapping, and partially non-overlapping neural structures for the motor control of these two classes of movement. In addition to this we report that responses in auditory cortices are distinct during the production of speech and non-speech vocalizations, suggesting that feedback control of speech and emotional vocalizations are distinct. These data provide novel evidence for the presence of dual pathways for the neural control of complex articulatory movements in humans. These findings are discussed in relation to the clinical and primate literature of vocal motor control.\n\nSignificance StatementThis work marks the first evidence in healthy humans for dual routes to vocal behaviors. Clinical evidence suggests that patients unable to produce speech may still be able to produce other vocal behaviors that employ the same set of effectors. Here we demonstrate that the production of different classes of vocal behavior is associated with overlapping and distinct networks of activity. Moreover, we show that auditory processing that occurs during the production of these movements may be distinct, suggesting that feedback may be used differently for these distinct classes of vocal movement.

neuroscience↗