Self-Regulation of Amygdala Activation by Unhappy Emotion Using Real-Time fMRI Neurofeedback with Autistic Spectrum Disorder -a Pilot Study-

Background: 　Functional magnetic resonance imaging (fMRI) has revolutionized the  study of the human brain. It was realized that as the fMRI technique has advanced the on-line processing of data would offer the possibility for immediate quality assurance of data and make real time fMRI (RtfMRI) data processing possible. RtfMRI enables a completely new type of experiment, whose designs and stimuli can adapt on-line to the participants' measured brain activity. Neurofeedback is a specific form of biofeedback, which feeds back information about brain activity to allow for training of voluntary regulation of brain activity. In this neurofeedback experiments using RtfMRI participants learn to control brain activity by contingent feedback of measures of the brain activity. In recent years, neurofeedback using RtfMRI applications have gained significant interest enabling potential clinical applications. It would allow us ask questions such as what areas are accessible to neurofeedback, how self-regulation is mediated, how  self-regulation can be learned, what kind of behavioral effects can be induced and how  specific they would be. This study addresses the potential clinical application of neurofeedback using RtfMTI for autistic spectrum disorder (ASD) in order to learn how to regulate their emotion.

Objectives: The current study investigates whether RtfMRI is a useful method for ASD patients to learn how to regulate unhappy emotion.

Methods: The ASD subjects (mean age: 18.6 years-old ) were enrolled in the study and had independent clinical diagnosis of Asperger’s disorder according to DSM-IV-TR. At first subjects were trained to down-regulate brain areas responsive  to unhappy emotions evoked by showing unhappy pictures. A target area was identified by the contrast between responses to unhappy and neutral pictures in a localizer scan to ensure that an area involved in unhappy emotion processing was selected. In the localizer scan, we assessed brain responses to unhappy and neutral pictures by presenting a set of pictures of the same emotion category. We used pictures from the IAPS. After subjects were trained with a set of pictures, subjects were instructed to downregulate the brain area responsive to words such as “ unhappy” and “neutral” in stead of the pictures. For the neurofeedback, a continuous signal from the target area  was displayed using the picture of a thermometer whose dial indicated the amplitude of the fMRI signal in the target area. Changes in the amplitude were indicated as the percent of signal change. Patients were not given any specific instructions about strategy. We acquired fMRI data on a 3 Tesla Siemens Trio-Tim.

Results: Subjects were able to decrease activation in amygdala during downgulation periods of the neurofeedback scans. Analysis of the contrast between conditions with word “unhappy” and “neutral” in the localizer scans yielded activation in the MPFC/insula. This indicate decreased activation in amygdala with nuerofeedback might be associated with MPFC/insula.

Conclusions: The pilot study shows ASD subjects were able to regulate amygdala activity and neurofeedback using RtfMRI for ASD has potential usefulness. Further studies are required to confirm the effectiveness of neurofeedback using RtfMRI for ASD.