The premise of the new research, which comes from University of Southern California, is that advances with neural decoding technology could allow for the prediction of mood in order to tackle affective disorders. The technology uses intracranial electrodes in order to capture neural signals in their brain and then to translate the resultant patterns into an assessment of the mood of a patient. This is based on the assessment of underlying neural signals.
Affective disorders
The pattern of affective disorders refers to a set of psychiatric disorders (sometimes simply called mood disorders). The primary types of affective disorders are depression, bipolar disorder, and anxiety disorder. Symptoms differ according to individuals and these range from mild to severe. With depression, symptoms of major depression are feelings of sadness, loss of interest in typically pleasurable activities (anhedonia), changes in appetite and sleep, loss of energy, and problems with concentration and decision-making.
Another related area is epilepsy, and patients with epilepsy were assessed in the trial. The term ‘epilepsy’ is itself a reference to a group of neurological disorders characterized by epileptic seizures. An epileptic seizure is a brief episode of signs or symptoms due to abnormally excessive or synchronous neuronal activity in the brain.
Running the experiment
To test out the theory, the researchers developed a modeling framework designed to decode mood state variations from multi-site intracranial recordings. The electrodes were fitted at the selected locations for seven human subjects who had been diagnosed with epilepsy. Prior to the study, each subject had self-reported their mood state at defined intervals across multiple days. This was through completing a 24-question self-report. With the report, the subjects were requested to ‘rate how you feel’ on a continuum of negative to positive mood pairs.
The most important data was obtained from the limbic regions of the brain. The limbic system supports a variety of functions including emotion, behavior, motivation, long-term memory, and olfaction (a chemoreception that forms the sense of smell). The analysis of the data captured by the electrodes showed that it was possible to demonstrate how mood state variations captured across time can be successfully decoded from neural activity. This was through the researchers using the recorded signals and correlating these with the questionnaire results.
The implications of the research are that the ability to decode mood state across time from an assessment of neural activity could lead to the development of a closed-loop system which could be used to treat a range of neuropsychiatric disorders. This would take the form of delivering electrical stimulation to the brain at the right moment in order to regulate unhealthy, debilitating extremes of emotion.
As lead researcher Maryam Shanechi told Future Science: “Our goal is to create a technology that helps clinicians obtain a more accurate map of what is happening in a depressed brain at a particular moment in time.”
Research paper
The new research has been published in the journal Nature Biotechnology. The research paper is titled “Mood variations decoded from multi-site intracranial human brain activity.”
Essential Science
This article is part of Digital Journal’s regular Essential Science columns. Each week Tim Sandle explores a topical and important scientific issue. Last week we reviewed how scientists are testing out machine learning algorithms which can make a prediction about yeast metabolism based on an assessment of the yeast protein content. This insight should aid scientists to personalize treatments for metabolic disorder patients.
The week before we considered a development with seismology. This was with a new machine learning approach that can help to predict where aftershocks, following an earthquake, are likely to occur. Aftershocks can often be of the same severity as the original earthquake.
