In 2012, Robert (Bob) Knight, Brian Pasley and team published findings in PLOS Biology revealing neural encoding mechanisms of speech acoustic parameters in the higher order human auditory cortex. The paper was greeted with media headlines declaring they had crossed scientific and ethical frontiers into “mind reading.”
At the time, Bob acknowledged the breakthrough but offered a different view of its implications, saying, “This is huge for patients who have damage to their speech mechanisms because of a stroke or Lou Gehrig’s disease and can’t speak,” he said. “If you could eventually reconstruct imagined conversations from brain activity, thousands of people could benefit.”
For a Jan, 2012 PLOS Biology podcast interview with Bob Knight and Brian Pasley on this paper,click here.
What follows is a new (June 2014) PLOS Neuro Q&A with the authors to discuss new areas of research the Knight Lab has tackled since 2012, including their work on the BRAIN Initiative.
1) How did your research on the human auditory cortex lead to your lab’s current focus on contributions of the prefrontal cortex to human cognitive behavior and mental illness?
My main research focus has always been on the role of human prefrontal cortex in cognition broadly defined. Main areas of interest have been attention, working memory, decision making etc. We started doing ECoG recording in 2003 and it became evident that we obtained nice auditory coverage in 70-80% of cases. Given the brain coverage, we began a series of auditory studies. The speech reconstruction work followed from this and Brian [Pasley] took the lead. I like this transitional area of research since I am a neurologist and the potential neuroprosthetic applications are exciting.
In this regard, we have some new data showing we can decode imagined speech. The effect is not as reliable as the initial PLOS paper but we are doing a series of studies to see if we can improve the results.
2) What were the key imaging technologies, research methods and engineering advances that have enabled this work?
Speech is a stimulus that varies rapidly in time and frequency. So the high spatial and temporal resolution of ECoG recordings in auditory cortices were critical for this work. Direct ECoG recordings allowed us to apply powerful statistical modeling techniques, originally developed from single unit animal studies, to investigate speech representation in the human brain.
3) What is your specific focus within the larger CNEP undertaking with UCSF?
1. Imagined speech reconstruction aimed at development of an implanted neuroprosthetic device for speech disorders
2. We were recently awarded a DARAP OBAMA Brain Initiative grant [in which] I am charged with figuring out the reward/decision making circuits that have gone awry in targeted psychiatric disorders. I am doing intracranial recording in our epilepsy patients to assess the contributions of lateral and orbital PFC and cingulate in reward/decision making. The long term goal is to develop novel intracranial stimulation methods (think DBS and Parkinson’s) to treat refractory depression and other psych disorders. Wish us luck, we will need it!
4) What other areas of brain mapping and cognitive neuroscience research do you find most exciting today?
Mine, just kidding!
I think the nature of the neural code remains the Rosetta Stone of human and animal neuroscience. Another very critical and poorly explored area is neuroplasticity. Lot’s of hand waving, very little robust data, very crucial for patients.
For more on the UC Berkeley/ UC San Francisco CNEP research agenda for the BRAIN Initiative, click here
If you have questions or comments for Bob Knight and Brian Pasley about their research or methods, please use this form to share your thoughts with the PLOS Neuroscience Community. Bob and Brian have agreed to respond here as their time permits over the next days and weeks.