May 16, 2015

Don't Don't Break the Chain... Make the Chain!

Make the chain!
(source: Wikipedia)
One of the worst productivity tips I've ever heard came from Jerry Seinfeld.  Here's the "tip": never skip a day of work.  Specifically, as a comedian-writer Seinfeld's goal was to write as much as possible.  To this end, he marked a calendar with a big red "X" for every day he worked.  After a few days of this, a chain of X's would form.  Then came the productivity goal: "Don't break the chain."

I've tried this trick for some activities and I've found it to be a terrible piece of advice.  Here's why: I always failed.  Eventually, I broke the chain.  Eventually, life reared up and made it impossible to stick with a habit.  Something always comes up: illness, events, deadlines, fatigue.  And guess what?  Failing sucks!

The problem with "Don't Break the Chain" is that it is an avoidance goal: it frames a goal or habit as something not to do.  The problem with this is that doesn't describe the behavior that one should be doing and leaves too much room for failure, as I've described.

Instead, I propose the following counter to Seinfeld's tip: "Make a chain."  Track every time you do something.  Make a check mark. Stick a sticker.  Put a coin in a jar. Whatever.  Just give yourself credit for every time you do it right and keep track of how many times you did it.  In this case, it doesn't matter if you skip a day.  Pick up the next day, or the next.  Or the next week.  Small failures don't matter if your goal is to make a huge chain; just pick up where you left off.

This has worked great for me while I've taken up running.  At first, I got all bummed out when I missed a day of my running plan: I broke the chain.  But then I reframed the goal as "make the chain" and now I can never fail.  If I miss a day, I just pick up where I left off.  Each workout I complete, I cross it off.  I see a permanent record of my progress that cannot be taken away.

Stay happy :)

May 11, 2015

The Psychology of Anthropomorphism

Today's Guest Post is authored by Mowaffak Allaham.  Mowaffak is a graduate student at GMU and a research assistant at the GMU Social Robotics Lab. Follow him on twitter at @mowaffakallaham.

Author: Mowaffak Allaham
Psychologists have identified the ability to perceive the minds of others as necessary for meaningful social interactions. Ongoing research is trying to determine factors that underpin mind perception, as this ability not only allows us to perceive the mind of a fellow human, but also to perceive it in nonhuman objects or agents. This tendency to imbue the real, or imagined, behavior of nonhuman agents with humanlike characteristics, motivations, intentions, or emotions [1] is called anthropomorphism.

A critical prerequisite to understanding the minds of other humans is to attribute the presence of mental states to their minds in the first place – intentions, desires, and beliefs. During the process of anthropomorphism, the attribution of mental states can even be applied to non-human objects or agents (e.g: 3D avatars or robots).  In a classic experiment exploring this phenomenon, Fritz Heider and Marianne Simmel [2] presented participants with a video of two animated triangles either chasing or hiding from one another.

This study demonstrated our innate tendency to attribute personality traits, and therefore a mind, even for simple, geometric shapes! Since then, anthropomorphism has intrigued many psychologists, and more recently neuroscientists, as a window into the cognitive mechanisms that drive our perceptions of the mental states in others.

Interestingly, one study found that an absence of social connections increased the tendency to anthropomorphize, presumably to satisfy our motivation for social connection.  In contrast, people with a strong sense of social connection were less likely to anthropomorphize non-human agents.

Research on anthropomorphism has expanded beyond the confines of psychology, reaching newly emerging fields like human-robot interaction. Computer scientists and roboticists are actively exploring the factors that influence our perception of robots.

Along these lines, scientists at the Robotics Institute at Carnegie Mellon University have proposed six design suggestions for a humanoid robotic head [1] to support the perception of humanness in robots. Further, these researchers have isolated some facial features in particular, such as eyes, nose, and eyebrows, as major contributors to a robot’s humanness. However, even robots that do not include all of these features, like Kismet at MIT [3], are sufficient for our minds to anthropomorphize and treat them in a very human-like way.

There is no doubt that robots are becoming more present in our lives, but what are the psychological implications of this new technology? Earlier this year Boston Dynamics revealed a video demonstrating their new robot “Spot”. This autonomous robot has four hydraulic legs and a sensor head to help it move across rough terrain. Although Spot’s appearance was quite robotic, many people condemned the act of kicking it during the recorded video demonstration. Some took it further to initiate a campaign to stop robot abuse. Interestingly, such reactions can inform us how people perceived Spot to have a mind similar to that of humans, therefore a feeling of pain, despite its obvious animalistic embodiment.

But what role does one’s belief, or knowledge, of a specific agent play in anthropomorphizing it? Noel Sharkey, a professor of artificial intelligence and robotics at the University of Sheffield, UK, has told CNN that for him, as a roboticist, kicking Spot was “quite an impressive test” since usually kicking a robot will knock it over. Was his prior knowledge of artificial intelligence sufficient to allow Sharkey to perceive Spot as a mindless agent?  His attitude was in contrast with those who perceived Spot, a robot without a head, as a mindful robot that feels pain despite lacking even the basic characteristics of an animal.  Nuances like these are essential to our understanding of how we anthropomorphize others and require greater understanding if we are to improve human-robot interactions. Knowing more about the cognitive, or neurological, process of anthropomorphism could assist computer scientists and roboticists to reverse engineer and implement the underlying principles in future caregiver robots, for example, improving interactions with patients. In other words, cracking the mechanism that underlies anthropomorphism could bring us closer to having robots that read, and help, the minds of others.


[1] DiSalvo, Carl F., et al. "All robots are not created equal: the design and perception of humanoid robot heads." Proceedings of the 4th conference on Designing interactive systems: processes, practices, methods, and techniques. ACM, 2002.

[2] Animating Anthropomorphism: Giving Minds To Geometric Shapes Scientific American.

[3] Breazeal, Cynthia. "Toward sociable robots." Robotics and autonomous systems 42.3 (2003): 167-175.

Further Readings:

Epley, Nicholas, et al. "When we need a human: Motivational determinants of anthropomorphism." Social Cognition 26.2 (2008): 143-155.

Epley, Nicholas, Adam Waytz, and John T. Cacioppo. "On seeing human: a three-factor theory of anthropomorphism." Psychological review 114.4 (2007): 864.

April 25, 2015

Is Meditation Self-Help Bullshit?

Are you wasting your time,
meditating monkey?
A recent article by Virginia Heffernan in the New York Time Magazine is excoriating the gradual westernization of mindfulness meditation, demonizing this trend as somehow running counter to the essence of this ancient practice.  I think at the heart of this article is a desire to protect people from self-help snake oil but there is a palpable vibe in the article of an anti-self-help bias that is unfortunate.

As I've discussed before, our attitudes about change influence our ability to change.  So, while I agree with Ms. Heffernan that self-help advice should be evaluated critically, binning the entire self-help movement as bullshit isn't helping anyone either.

In this context, I can't help but reevaluate the purpose of mindfulness meditation (and mindfulness, in general).  Is mindfulness mediation useful? 

As someone who has practiced mindfulness meditation as an attempt to manage stress, I have concluded that meditation is simply a concerted effort to implement a reappraisal of bad thoughts.  Specifically, it has been suggested that rumination, or the endless replay of negative thinking, may contribute to depression.  Cognitive reappraisal is a well-known approach for dealing with a number of negative or disruptive thoughts and meditation is just a practiced form of this.  In the mindfulness meditation style I have tried, namely Mindfulness-Based Stress Reduction popularized by Jon Kabat-Zinn, one reappraises negative thoughts as "thoughts", taking a meta-level view of them and partitioning them as something distinct from our experiencing mind.

Personally, this makes sense to me.  Just as I wouldn't accept the self-help advice of some rando, I am not going to trust that my automatic catastrophizing about the world is based on fact.  During mindfulness meditation, I am taking a skeptical stance toward my own worries and recognizing them for what they are: worries, not reality.  Ultimately, the proof is in the pudding when it comes to the value of mindfulness meditation.  If it helps someone cope with the challenges of life, then great.  We should all be active participants in our own mental health, experimenting with approaches until we get results.

In this way, stress management is like exercise.  Is one form of exercise better than another?  Is meditation better than a book club?  The answer is: it depends.  It depends on who is doing it, whether they enjoy it and whether they stick with it.  If the answer to these questions is "yes", then the long term outcome is likely to be positive.

April 09, 2015

My Stress Portfolio

What's your stress tolerance? (Credit: Mike Richey)
I'm no investor.  Far from it.  I've bought maybe 10 stocks in my lifetime and I recall the best performer losing half it's value.  I'm also no sucker, so I learned from my mistakes and quickly outsourced my meager investment portfolio to the professionals and haven't looked back.

Now, when I talk about "investment professionals", I should clarify.  I really don't trust the stock-picking ability of anyone since the financial markets are complicated and I doubt that anyone can predict how anything that complicated will behave.  Fortunately, I'm not alone in this opinion and a subset of the investment community relies on the principle of diversification of risk as a bedrock strategy for coping with the uncertainty of the markets.  Through diversification, an investor can rely on historical information about the best performing investment types without putting all her eggs in a single proverbial basket.

Now let's shift gears to stress.  Life activities and events are like investments.  We are investing our time and attention in activities that may bring us happiness, pleasure, grief, or stress.  The payouts or losses may be on the timescales of days, to weeks, to years.  As an investor of our time and attention, we must weigh the probabilities of gains with the risks of losses: driving hard at work could bring promotions at the cost of personal relationships, for example.

I propose the concept of a Stress Portfolio as a strategy to balance our activities to allow both personal growth while managing the risks that come with stress, like poor health or burnout.  My Stress Portfolio is a introspective sense of the balance of five components of my reaction to daily life: distress, eustress, flow, relaxation, and boredom.  I think of each of these as representing a decreasing level of stress one might experience.  As events happen in life both within or outside my control, I can reevaluate the distribution of my stress portfolio and "rebalance" my intensity to compensate.

What is the right distribution of stress?  I think the answer to that question will vary by the individual. For example, some people are more resilient than others and may benefit from a greater allocation of activities in the distress or eustress categories.  This is an interesting area for research as proper allocation of stress will determine the degree of success an individual can achieve.  To much stress and a person may crumble and give up.  Too little stress and a person is not challenged and may not rise to their full potential.

As a general rule, I think most of our time should be spent in a state of relaxation or flow with a decent amount of time in a state of eustress (exciting or exhilarating situations). I also don't advocate total avoidance of stressful situations or boredom (the extremes).  These act as high and low intensity activities that can permit personal growth and recovery, respectively.

Until more research is done to determine the right allocation, all we can do is look within and ask ourselves what our portfolio should look like.  Just like an investor must evaluate her risk tolerance, an individual must evaluate her stress tolerance to find the right allocation and maximize performance over time.

Stay happy!

April 01, 2015

Useful Data

But is the information useful, monkey? Is it?!
In a recent attempt to "cheat" with science while creating a March Madness bracket, I rediscovered the online data-geek candy store, brain child of renowned statistician Nate Silver.  In another incredible display of sexy stats, Silver and Company have assembled a robust statistical model of the NCAA men's basketball tournament, adding evidence to the thesis that the geeks shall inherit the earth.

While the information on FiveThiryEight is beautifully presented and likely as accurate as one will get, I began to wonder: is this useful? More broadly, what makes information useful?

I don't want to get too hung up on what "useful" means, but for my purposes I'd like to define useful as enabling better performance: greater accuracy, greater speed, or higher success rates in some activity.  What feature of data would make it useful?

Here's my opinion: for information to be useful, it must be actionable.  In other words, for information to enable better performance or higher success rates it must inform what actions should, or should not be performed to improve accuracy.

A great example of this type of information for me has been heart rate data during exercise.  When I run according to my heart rate training plan, I know that I'm working too hard on my easy day when my heart rate goes above some threshold.  This is actionable information.  My heart rate is too high so I change my behavior and slow down.

For the FiveThirtyEighty March Madness predictions, the information is sort of actionable.  The FiveThirtyEight bracket is structured as probabilities of a team winning at each stage of the tournament.  This information is useful if I'm betting on a game (i.e. which team is likely to win) but isn't useful if I'm trying to make a bracket (i.e. which teams are most likely to be in each slot of the bracket).

The lesson here is that the structure of the information should match the decision to be made.  In the example of my hear rate data, my current heart rate is only useful in the context of a threshold.  I must know how my current data point relates to some useful scale.  Only then can I take action to bring my individual measure back into range.

However, generating actionable data is incredibly complicated because it requires a solid understanding of the mechanisms that explain a phenomenon.  In the case of heart rate training, heart rate is a well-established proxy for intensity and systematically modulating intensity is important to balance improvements in fitness against risk for injury.   Creating actionable data is also complicated by the need to understand how it will be used (i.e. betting on a game vs. making a bracket).  For these reasons, generating data without a solid theory to back up action is no better than rock collecting.

A related piece on FiveThirtyEight highlights an interview with White House Chief Data Scientist, Dr. D.J. Patel.   A topic of discussion that caught my attention was that of "Data Products" which Dr. Patel explains as "How do you use data to do something really beneficial?"  The true obstacle to creating powerful data products (the dream of Big Data) isn't access to data or data processing tools as these are now as ubiquitous  as the internet and the personal computer, respectively.  Instead, the obstacle to data products, or "useful data" as I'm calling it, is creating a solid theory about the mechanism driving a phenomenon.  Without this understanding, data remains noise.

For these reasons, while I am as enamored by sexy data as much as the next dork, I am reminded about the need for good scientific theories that allow us to interpret and structure sexy data in a way that makes it useful.  That is the real challenge for data scientists in the age of big data.