Network-based fMRI-neurofeedback training applied to sustained attention

• Ano de defesa: 2018
• Autor(a) principal: Pamplona, Gustavo Santo Pedro
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Biblioteca Digitais de Teses e Dissertações da USP
• Palavras-chave em Português: Alterações comportamentais ; Atenção ; fMRI em tempo real ; Neurofeedback ; Rede de atenção sustentada ; Rede de modo padrão; Attention ; Behavioral changes ; Default Mode Network ; Neurofeedback ; Real-time fMRI ; Sustained attention network
• Link de acesso: http://www.teses.usp.br/teses/disponiveis/59/59135/tde-31102018-081404/

Attention is a key mental function in everyday life, but unfortunately we easily get distracted. The brain correlates underlying sustained attention, the so-called sustained attention network (SAN), have been well identified, as have the brain correlates underlying mind-wandering, the so-called default mode network (DMN). Nevertheless, even though we know about the underlying brain processes, this knowledge has not yet been translated in advanced brain-based attention training protocols. Here we proposed to use a novel brain imaging technique based on real-time functional magnetic resonance imaging (fMRI) to provide individuals with information about ongoing levels of activity in the attention and the default mode networks. To the best of our knowledge, this is the first study to show that, with the help of that fMRI-neurofeedback, individuals can learn how to improve controlling of, at the same time, SAN activation and DMN deactivation. This learning process was explained mainly in terms of DMN deactivation. Behavioral effects were observed when separating a group with the best learners in an overall measure of attention and specifically in the task-switching ability, controlled by a test-retest group performing the same behavioral tests battery. Neurofeedback-induced functional connectivity changes were also observed in multiple brain regions positively and negatively related to attention. Although the behavioral effects were no longer present two months after training, participants still held the learned ability of controlling self-regulation of the concerned networks. This approach potentially provides a non-invasive and non-pharmacological tool to deliver general enhancements in the attention ability for healthy subjects and it can be potentially beneficial to many neurological and psychiatric patients. We also show in this thesis compelling evidence that brain regions definition and other experimental parameters are crucial for inducing learning of self-regulation via fMRI-neurofeedback, in a similar study also considering differential signal of attention-related competitive networks. We finally present Personode, a useful, easy to use, and open access toolbox to neuroimaging researchers, for independent component analysis maps classification into canonical resting-state networks and regions-of--interest definition in individual and group levels. We also show that the toolbox leads to better results for task-induced activation and functional connectivity analyses.