Crowdsourcing the Classic Reframe

As I mentioned in my last post, I spent Mother’s Day participating in a workshop focused on technology and mental health. Specifically, innovation in the user experience (UX) with technology as it applies to identifying, intervening, and preventing mental health problems.

Since attending the workshop, one of the take-home experiences I’ve been exploring is the Koko app that crowdsources the classic counseling technique of reframing a client’s statement. Reframes are a therapeutic, conversational technique that involve reflecting what a client has said, but not in a mere “parroting” of the original statement, but in a way that helps the client see their situation from a different perspective, often one that is more “open” to change.

For example, a client might say, “I’m completely burned out by my current project at work which is a massive failure and something that I hate doing”. Therapists are trained to pickup on “absolute” or “extreme” terms, such as completelymassive and hate. While possibly accurate from a given perspective, they are also limiting and oppressive, such statements don’t afford many options. A reframe might be, “I can tell you’ve been working really hard and are struggling to make things come together. That’s limiting the satisfaction you get out of your work“. This reframe validates the client’s effort and emotional experience and highlights an area where she is not getting her needs met. A reframe “works” when the client identifies with it, and they experience an emotional and cognitive shift. If not, the conversation must continue.

A tiny fraction of people on earth have access to trained counselors, and even those individuals rarely use them. There are many barriers to engaging in counseling, and until the advent of recent technological advances, we had pretty limited capacity to make services more available. We could basically train more people and attempt to reduce transportation and time barriers. But over several decades, we haven’t gotten very far with that approach. The internet offered a moderate advantage by reducing the barriers of time and travel, but initial applications still relied on expert counselors to engage with clients. “Mechanized” interventions began to emerge, but these were along the lines of self-help books, there was no dynamic tailoring of client needs with therapeutic responses.

Enter crowdsourcing and artificial intelligence. Crowdsourcing is essentially a disruptive use of technology which harnesses the wisdom of the collective rather than reliance on individual experts. Perhaps the most well known, and one of the first such applications, is Wikipedia. When Wikipedia first started getting attention, most people could not believe that an “open-source” encyclopedia could produce accurate and quality information. Today we know better, and the encyclopedia as I knew it growing up, has gone the way of land-line telephones.

But Wikipedia does not dispense with expert individuals. It is a curated repository of information and there have been several generations of innovation with respect to how Wikipedia manages content in order to insure quality and attempt to avoid major controversy.

As Rob Morris explained at the workshop, Koko started out using something similar to the Wikipedia model, insomuch as users of Koko could post their troubling thoughts and the Koko community (anyone interested in signing up), would respond to posts with their attempt at reframing the original post in such a way that validated the poster and helped them see their situation from a different perspective. As I mentioned in my previous post, the model was also very similar to the StackOverflow platform in which individuals post problems they are having related to programming or statistics or a host of technology related issues, and community members attempt to help them out. Like Wikipedia, the content is monitored by individuals who have earned the privilege of content management through their judicious participation in the community.

The challenge of course is how do you build up a community of experts who can curate the content? To start with, Rob and one of his partners were curating responses to user posts. They had to read every response and determine if it was appropriate or not. Mostly, I believe they were trying to just weed out the sometimes cynical, critical, or judgmental responses people would propose.This very quickly became a major bottleneck and in turn became the inspiration for harnessing the power of artificial intelligence to scale the platform to be able to support thousands (hundreds of thousands of posts). It’s my understanding, that they are currently using IBM’s Watson to do two important tasks, 1) curate responses and filter out harmful content; and 2) to abstract the user posts into pithy summaries. It appears that the first one is working well, I haven’t seen a response come through that I felt was harmful or negative. Granted, my observation count is small and I imagine there aren’t too many of these. The second goal, the abstraction definitely has room for improvement. Many are spot on, but the algorithm currently reads too much into statements and produces pithy statements that overstep their bounds. However, I don’t see this as a major problem so long as the community focuses on the original post and does not rely on the pithy abstraction.

But, don’t take my word for it. Download the app and give it a try!

Advertisements

Mother’s Day on Mental Health Technology Steroids

I spent Mother’s day attending an all day workshop hosted by the ACM Special Interest Group – Computer Human Interaction on Computing and Mental Health as a part of the CHI-2016 annual convention.

The workshop brought ~90 individuals interested enough in the application of technology to promote wellness, prevent illness, and treat mental disorders to submit proposals and then travel to San Jose to attend an all day workshop on Mother’s Day.

This was truly a multidisciplinary event with social workers, counseling and clinical psychologists, psychiatrists, computer scientists, electrical engineers, interface designers, and several more disciplines I’m sure. I have to really give credit to the computer scientists who had the vision and generosity to bring together these often silo’d worlds (Rafael Calvo, Rosalind Picard, Karthik Dinkier and Patty Maes). Check out their work and the lasting impact they are having on mental wellness will be more than obvious.

The format of the workshop was a mix of 20-minute presentations by folks like Tom Insel, Eric Horvitz, Matthew Nock and more, interspersed with 5-minute project presentations, and a poster session (fortunately the 5-minute presenters also had posters). The event ended with a substantial Q&A with the panel of keynote speakers.

Although the “30,000 foot” perspective offered by industry luminaries were on target and offered inspiration motivation, it was clearly not needed by this passionate group.

I do want to give special mention to two of the keynote speakers, Helen Christensen,  Chief Scientist of Black Dog Institute  and Robert Morris founder of Koko. Many of you have probably seen the “black dog of depression” video, a powerful metaphor based explanation of the lived experience of depression. What you may not know is that the black dog video one of dozens of technology-based mental health promotion, prevention and treatment projects coming out of the prolific labs at Black Dog Institute. A similarly inspiring presentation was given by Rob showcasing the crowd sourced CBT intervention platform for depression he cofounded. In the past few years I’ve been dumb-struck by the immense benefit that crowd sourced platforms for technological knowledge have contributed to my learning process (e.g. stackoverflow.com and crossvalidated.com). Koko harnesses that model but applies it to principles of cognitive therapy for depression in the most democratic and open sourced delivery system imaginable. I draw special attention to Helen and Rob’s work because I believe it represents some of the best-of-the-best bench to living room translational science I know of.

Although the work of Helen and Rob and their respective organizations represent the cutting edge, it was the work presented by the 40 some participants who represent that Gatesian horizon of ten year innovation that is likely to radically transform the benefit of technology on our wellbeing.

You can see short papers describing all of the presentations here.

One of the special features of much of the work presented that really impressed me, having been a frustrated mental health services researcher for over ten years, is the incredible ability of many of these technologies to be delivered at a global scale in months or years rather than the traditional health innovation cycle of decades. We are losing the shackles of glacial implementation timeframes and moving towards an era of amazingly fast cycles of innovation, development, and dissemination. No longer is innovation the captive  of government and big-medicine beaurocracy these technologies and innovations go straight to the source, the user. I have no fantasy that this change will be pain free, but it is coming and if the field of psychology can get on board fast enough, we may be able to contribute to our dreams of massive reductions in suffering  caused by mental illness and societies’ ill-equipped ability to respond to it.

And finally, a shout out to my awesome research team 2015-2016 who helped me  conduct some extremely rewarding research this year and thus to participate in the CHI-2016 Workshop on Mental Health – Melanie Cadet, Madeline Brodt, Bonnie Andrews, Sam Gable, and Meredith Maroney.

 

#EmotionLab16 Hackathon, wow!

This weekend I participated in my first high-tech hackathon. I qualify it as a first because last December I participated in a hackathon at the VA focused on innovation in health care delivery. However, these two events were very different. The general process was similar, but the VA event drew very few (zero?) engineers/coders, so we spent the day thinking about ideas without doing any development. This makes sense if you are focusing on innovation in health services.

But, if you are focusing applications for cutting edge technology you need coders. Affectiva, the primary sponsor of the event is a magnet for coders, and they showed up in numbers. The other thing you need is cutting edge software to infuse applications with new possibilities. Affectiva’s new API for facial emotion coding was the big draw, but there were other amazing tools available, including BeyondVerbal‘s API for coding emotion from audio data, and BrainPower brought their amazing Google Glass API for integrating Affectiva into Google Glass applications. Another cool device was Pavlok’s haptic feedback device, check it out, it’s shocking! There were other cool tools as well, but these were the products I focused on using.

For the uninitiated, a hackathon is an event where a group of diverse individuals come together for a day or a weekend, form teams (usually on the spot with other participants) and knuckle down to develop an idea into a prototype, and then pitch their idea to a panel of judges to win prizes for creativity, business potential, etc.

I attended with hopes of forming a team to develop an app for assessing high-risk patients’ mood periodically throughout the day by having them talk about their day rather than to fill in a traditional self-report questionnaire. I was open working on any team really, but my hopes were fulfilled. A team formed around my idea (Skyler Lauren, Annie Tuan, and Kay Corry Aubrey) that worked amazingly hard to produce a working iPhone app that can analyze facial and verbal affect and transmit that data to a server for in-depth analysis!

I have to admit, I harbored a delusion that our app was going to be the coolest thing at the event; I was totally blown away by the creative output that occurred. There were ten teams and the projects were mind-blowing.  Unfortunately, I can’t describe these products as many if not all are initial business ideas and very cutting edge. Almost all of the apps were focused on helping individuals overcome limitations in information processing or enhance human performance. There were also some fun applications that I look forward to seeing on my mobile device soon.

I believe there is immense potential for psychologists to advance their work through this type of event. I would love to do more of these and fewer academic conferences (or, maybe what I really want is for the conferences I attend to incorporate hackathons!)

You can learn more all over the internet, if you are on twitter, search for #EmotionLab16 to see lot’s of photos from this event.

Does marijuana inhibit emotion recognition? I’m not holding my breath just yet.

Yesterday I introduced a new focus to this blog, namely a goal to write about the massive impact technology is having on the study of human behavior. As I noted, my motivation comes in part from my concern that traditional academic psychology fields (specifically my home discipline of Counseling Psychology) are way behind the curve on these advances and may likely become obsolete if they aren’t a part of the impending revolution.

Today my attention was directed to a recent article published in PLOS ONE titled, An Event-Related Potential Study on the Effects of Cannabis on Emotion ProcessingFirst, I want to give a shout out to PLOS ONE for being an open access journal that does not restrict the number of pages, figures, or tables you can include in your paper, and advocates strongly for data-sharing. I also want to express my appreciation to the authors of this study who have provided us with a wealth of new data and knowledge regarding emotion-related brain function and the potential impact on such by chronic use of marijuana.

This study has some great strengths, and I particularly appreciate their focus on constructivist models of emotion processing over “locationist” models. But what I really want to call out, is that this study highlights a critical challenge to the application of biometric measurement to human “behavior”. The authors employed two distinct measurement methods to evaluating the impact of marijuana use on emotion processing, one behavioral and the other biometric. The behavioral measure used a long-standing methodology of measuring reaction times to various stimuli. The biometric measure utilized EEG technology to detect specific brain activity. The results were conflicting, that is the behavioral measure detected no differences between marijuana users compared to non-users, whereas the EEG measure did detect differences.

So what’s going on and what is important? Based on the excellent introduction to the study, there is strong evidence that the EEG measure used is related to emotional information processing, yet the EEG results were not consistent with the behavioral measure. My conclusion is that we don’t know, and whatever impact marijuana is having on these individuals, it isn’t impacting their ability to respond to emotion related stimuli. Well, let me qualify that last statement, it isn’t impacting their reaction times as measured by how long it takes to strike a computer keyboard to categorize stimuli. It may very well be impacting important emotional processing and it’s just that this particular study didn’t capture how that impact manifests in overt human behavior or experience.

Human behavior is complex and we have a long way to go before we unravel age-old mysteries. Technology is going to help us out, but along the way, we have a lot of work to do before we can tie small and specific biometric signals to behaviors that impact our quality of life. Reductionists beware and Luddites, wake up.

One more shout out to Rana el Kaliouby (@kaliouby) for tweeting about the article discussed today.

Affective computing and the future of psychology

A little over a year ago, I read an article in the New Yorker about the field of affective computing titled, “We Know How You Feel.” That story has had a profound effect on my work as a suicidologist and has changed my perspective on the role of technology within Counseling and Clinical Psychology.

Heretofore I have considered myself an early adapter, my first career was in computer programming and I got my first personal computer in 1983; which was followed by Newtons, Palm Pilots, iPhones, etc. But when I entered graduate school in Counseling Psychology in 1999, I put aside my interest in using technology to solve scientific problems and dove into the study of human behavior from a traditional framework. I knew artificial intelligence was advancing and that robotics was on the go, but I didn’t see how close these fields were to revolutionizing our understanding of computer-human interaction and the ability of computers to help us understand and influence human behavior.

It wasn’t until I moved to Boston to become faculty at UMass Boston and I re-engaged my interest in programming (for the sake of data management and analysis) that I started catching up on what had happened in the last 15 years. I was truly shocked to discover an entire field, affective computing had emerged, and was producing tools and knowledge at a level that far exceeds what can be done with the traditional tools of psychology.

Although there have long been siren calls regarding change that was bound to happen in the future, I am acutely aware that the future is now with respect to computer-human interaction and I am deeply concerned that traditional academic programs in Counseling Psychology are already archaic and will soon be obsolete. I know that sounds extreme, but check out some of the links I’ve provided below, and stay tuned as I continue to write about the amazing opportunities that are waiting for us to take advantage of…

  • Fully automated emotion sensing and analytics available on your smart phone by Affectiva.
  • Mobile infrared eye-tracking technology easily integrated with multiple biometric modalities by iMotions
  • Virtual human platforms that sense and respond to emotional dialogue with real humans by the USC Institute for Creative Technologies.

The Power of Stories

Researchers are trained not to trust anecdotal evidence, the information that comes from a single observation. As a psychologist, with my researcher hat on, this means that I’m often discounting the significance of a person’s story, their explanation of events, or their meaning making. Yet doing so, violates what politicians and business people have known forever, there is nothing more powerful and persuasive than a person’s story. It also violates what I, with my therapist hat on, know and do. This tension has been explored through many avenues, including the discipline of philosophy of science itself, yet the challenge remains. It is this very tension that brought my attention to the discourse of mental illness, for it is within this context that much of our language around a human behavior is formed.

One of the ways that I helped manage that tension last fall in the Abnormal Psychology course I taught was to balance research derived descriptions of mental illness with first person narratives. I found wealth of resources out there in the form of books, audio recordings, and videos. The national center for PTSD has a wonderful project called “about face” that chronicles the experiences of veterans with post traumatic stress disorder.

In today’s new york times, Well section, there is a story about a project coming out of Cincinnati Ohio that chronicles the lived experience of people with schizophrenia. The story highlights one of the most harmful aspects of schizophrenia, the social isolation that comes with it. This isolation is self-perpetuating, without knowing the experience of people with schizophrenia, we fear them and avoid them. I encourage you to go and listen.

A Psychiatrist Dabbles in the Contextual Aspect of Antidepressant Medication In College Aged Adults

In yesterday’s NYT, psychiatrist Doris Iarovici, writes as short reflective essay in the Well section, titled “The Antidepressant Generation”.

Dr. Iarovici explores the notion that perhaps the national standard of practice for prescribing antidepressants to young adults is not such a great idea. She explores two concerns: 1) could there be unknown, long term negative neurological consequences to being on antidepressants, and 2) are we socializing young people to medicate what are developmentally normal challenges? It’s a short essay, and reflects questions that have been raised by many for a long time already, but I’m happy to pause and reflect myself.

The question gets posed within the scientific paradigm in the following paragraph:

Are we using good scientific evidence to make decisions about keeping these young people on antidepressants? Or are we inadvertently teaching future generations to view themselves as too fragile to cope with the adversity that life invariably brings?

The sad thing about this article, which happens over and over again, is that there is no mention of the fact that psychotherapy is a perfectly viable, evidence based treatment for depression with substantially better evidence and scientific support than pharmacotherapy. By ignoring this option, the situation gets framed as a zero sum game in which either you suffer from depression or you bear the potential harm from antidepressants. It’s hard to understand how this gets left out, particularly when the main concerns that Dr. Iarovici discusses, neurological damage and failing to prepare young people for the psychological demands of life are directly alleviated with psychotherapy. In therapy there are no potential risks of neurological damage and regardless of the type of therapy one engages in, the end goal is improved psychosocial functioning through psychological and behavioral changes that the client incorporates into his or her life.

In this essay, the issue also gets framed in a double bind discourse. Either we have medicines that can fix the biological imbalance causing you to be depressed, OR you are too fragile a human being to handle life. Really, is that the best we can do in our common discourse about depression? Sadly, when framed this way people are going to prefer the biological model and take the risk; not doing so is to shame themselves. 

 

History of mental illness used to justify fatal shooting of homeless man

The Huffington Post featured a story coming out of Albequerque New Mexico in which a homeless man was fatally shot by local police after an apparent 3 hour standoff.  In the Huff piece, there is a video from the helmet of one of the officers involved in the incident, it clearly shows the police shooting the man in the back and then firing numerous rounds to assure that he was dead, with no intention of trying to save his life. 

The police force attempted to justify their use of force based on the man’s history of mental illness and violent behavior:

Boyd, who police said may have been a paranoid schizophrenic [sic], has a long criminal history. In the past, he allegedly attacked people with knives, box cutters, and his hands, and in 2010 broke a female officer’s nose, according to KOAT.

More disturbing than this cover article though is the referenced article at the ABC affiliate news station “KOAT” in Albequerque. The link above is set in the same way as it appears in the Huff article, “has a long criminal history”. In that article the portrayal of the victim, James Boyd, is 100% that of a violent, multiple offender with a deviant history. 

Sad discourse. There are many important social issues involved in this situation, I’m choosing to focus on the discourse related to the portrayal of the victim as it is the most relevant to the blog. Even if one takes the KOAT story literally and completely, it seems clear that James Boyd had a troubled life. My interpretation of the KOAT article is that it aims to justify the use of deadly force based on the impression that Mr. Boyd had engaged in strange, violent behavior in the past. Of course, the flip side is that our mental health and social service system failed this man, and our public safety system has been socialized into a mentality that it is okay to use excessive force provided the individual can be cast as a social deviant in a few paragraphs – and that serious mental illness is a justifiable way of dehumanizing an individual.

 

 

1 in 5 nursing home residents given antipsychotic medication – behavioral interventions largely ignored

The Boston Globe is running a series of articles related to the overuse of antipsychotic medication in nursing home residents. They report that nationally 1 in 5 nursing home residents receive antipsychotic medication and that in Massachusetts the number is close to 1 in 3.

Anybody that has had a loved one suffering from dementia is likely to be familiar with the pain of a deteriorating mind. Those who have been care givers, either personally or professionally can attest to the challenges of trying to “manage” a person suffering from dementia. Rates of depression and stress in caregivers has been well documented and it is a serious concern. So it’s not surprising that the medical field is responding by throwing antipsychotic medications at the problem – a quick fix to a difficult and emotionally painful challenge.

In the Globe series, there is one article that discusses behavioral interventions that have been shown to be highly effective at relieving anger, agitation, stress, and depression in dementia patients with what seems like common sense intervention – creating a peaceful, comforting, holistically healing environment. As the Globe article points out, these types of interventions are labor intensive and take time – they are not guaranteed to work in the same way a potent sedative will.

It’s frustrating to read the series and see how little attention is paid to alternative solutions. The energy goes into pointing fingers and assigning blame rather than championing better solutions. A quick response to a difficult and emotionally painful challenge.  

There are so many people who would absolutely love to have a job creating peaceful, healing environments for seniors – why are we instead further debilitating their minds with drugs?