Computational Linguistics and Clinical Psychology — From Linguistic Signal to Clinical Reality
A workshop held on June 27, 2014 in conjunction with the 2014 Conference of the Association for Computational Linguistics in Baltimore, Maryland, USA.
Motivation: Why a new workshop?
Mental health problems are among the most pressing challenges we face. The numbers in the U.S. alone are staggering: to cite just a few, between 1996 and 2011, annual expenditures on mental disorders rose from $35.2B to $113B, some 25 million American adults will have an episode of major depression this year, suicide is the third leading cause of death for people between 10 and 24 years old, and 89.3 million Americans live in federally-designated Mental Health Professional Shortage Areas. The importance of mental health as a problem space cannot be overstated.
For mental health professionals, language plays a central role in diagnosis. Indeed, many clinical instruments and neuropsychological assessments fundamentally rely on what is, in effect, manual coding of patient language. Applying language technology in this domain, e.g. in text- and speech-based assessment, could potentially have an enormous impact. For example, many individuals are motivated to underreport psychiatric symptoms (consider active duty soldiers, for example) or lack the self-awareness to report accurately (consider individuals involved in substance abuse who do not recognize their own addiction), and because many people — e.g. those without adequate insurance or in rural areas — cannot even obtain access to a clinician who is qualified to perform a psychological evaluation.
Bringing language technology to bear on these problems could potentially lead to inexpensive screening measures that could be administered by a wider array of healthcare professionals, which is particularly important since the majority of individuals who present with symptoms of mental health problems do so in a primary care physician’s office. Given the burden on primary care physicians to diagnose mental health disorders in very little time, the American Academy of Family Physicians has recognized the need for diagnostic resources for physicians that are “suited to the realities of their practice”.
Although automated language analysis connected with clinical conditions goes back at least as far as the 1990s , it has not been a major focus for computational linguistics compared with other application domains. However, very recently there has been noticable uptick in research activity on this topic. One recent shared task brings together research on the Big-5 personality traits, and another involved research on identification of emotion in suicide notes. Research has been done on language analysis in the context of autistic spectrum disorders, dementia, depression, post-partum depression, schizophrenia , general life satisfaction, and suicide risk. This increase in attention is consistent with, and gains power from, the recent rise in computational linguistics activity connected with computational social science more broadly. The bottom line: computational linguistics for mental health is a subfield that is ready to emerge, and we hope to help make that happen.
Workshop goals and topics of interest
With these motivations in mind, this workshop will focus on on language technology applications in mental health. We aim to bring together natural language processing (NLP) researchers with practicing clinicians, with the following four goals:
- To increase language technologists’ understanding of what mental health clinicians do and what their real needs are,
- To increase clinical practitioners’ understanding of what’s possible in language technology and what it might have to offer,
- To identify challenges that need to be overcome, and
- To formulate targets and priorities for near-term improvement of the practical state of the art.
In support of those goals, we are adopting a best practice from many social science conferences: instead of just the traditional presentation-plus-questions format, each paper will also have a clinically-oriented discussant, who will read the paper thoroughly in advance and briefly present prepared commentary.
In the context of the overall goals above, we are particularly interested in submissions that bear on issues like the following, relative to psychological conditions and neurological disorders:
- What features of language or speech could play a prominent role in diagnosis?
- What algorithms and forms of modeling are applicable?
- What kind of data exists or could be obtained?
- What practical or ethical issues require attention?
We aim to emerge from the workshop discussions with a well defined strategy for progress that is informed by both the ACL and clinically oriented participants. This could include, for example, identifying a specific topic, data, and task for a shared task; or formulating a plan for creating and sharing IRB application templates for natural language processing work in psychology; or identifying the top-level requirements for an natural language processing toolkit specifically devoted to practical issues in clinical psychology.
A key goal of this workshop is not only to bring together language technology researchers in this emerging field, but to foster the conversation with clinical practitioners, both at the workshop and when these papers are read in the future. For that reason, we are including practicing mental health clinicians on our program committee. The ability to communicate ideas, approaches, and results clearly to people who are not computational linguistics experts will be as important as the quality of the work itself.
With that in mind, we encourage you to include supplementary materials up to 2 pages in length containing technical details kept out of the body of the paper. These and any other supplementary materials, such as code or datasets, will be linked to via a URL in the final version of the accepted paper.
Submissions can contain up to 8 pages of content, plus references of any length, plus up to 2 pages of supplementary materials as described above. Supplementary materials should be self-contained (i.e. should start on a separate page). Papers must conform to the ACL 2014 submission format guidelines, as detailed under “Format” and “ACL 2014 Style Files” in the ACL 2014 Call for Papers. Submissions should be anonymous, i.e. please do not disclose in any way the identity of the author(s). Papers must be submitted using the START system at https://www.softconf.com/acl2014/CLPsych.
We will accept submissions for either oral or poster presentation.
- Philip Resnik, PhD, University of Maryland
- Rebecca Resnik, PsyD, Mindwell Psychology Bethesda
- Meg Mitchell, PhD, Microsoft Research
- To contact the organizers, please mail firstname.lastname@example.org.
- Steven Bedrick, Oregon Health and Science University
- Craig Bryan, University of Utah
- Jesus Caban, Walter Reed National Military Medical Center
- Munmun De Choudhury, Microsoft Research
- Michael Gamon, Microsoft Research
- Kristy Hollingshead, Department of Defense
- Arthur Horton, Psych Associates of Maryland
- Loring Ingraham, George Washington University
- Molly Ireland, Texas Tech
- Michal Kosinski, Cambridge University
- Antolin Llorente, University of Maryland Medical School
- Elmar Nöth, University of Erlangen-Nuremberg
- Serguei Pakhomov, University of Minnesota
- Emily Prud’hommeaux, University of Rochester
- Nan Bernstein Ratner, University of Maryland
- Ehud Reiter, University of Aberdeen
- Brian Roark, Google Research
- Andy Schwartz, University of Pennsylvania
- Kathy Seifert, Eastern Shore Psychological Services
- Thamar Solorio, Univ of Alabama at Birmingham
- David Stillwell, Cambridge University
- Paul Thompson, Dartmouth College
- Lyle Ungar, University of Pennsylvania
- Marilyn Walker, UC Santa Cruz
- Karin Scheetz Walsh, Children’s National Medical Center
- General information: https://clpsych.wordpress.com/
- Paper submissions: https://www.softconf.com/acl2014/CLPsych/
- ACL conference site: http://www.cs.jhu.edu/ACL2014
Submissions deadline : March 21, 2014
Notifications sent to authors: April 11, 2014
Final, camera-ready papers due: April 28, 2014
- Workshop in Baltimore: June 27, 2014