Friday, 2 January 2015

Musing--the problem with feedback

I plan a series of reflections, not so much on the Muse specifically, but on the issues that it and other similar devices raise.  This first musing is on the issue of feedback during meditation, especially auditory, which is used in the Calm app.

Ideally, there would be no need for feedback when we meditate.  We should be able to notice when our mind is focused and when it is distracted.  But this requires practice, and, even with practice, it is sometimes difficult to maintain the appropriate focus and awareness.  So this is where devices such as the Muse might prove helpful.

However, the biggest issue with using feedback technologies with meditation is that certain types of feedback can actually interfere with the activity they are designed to assist.

Let's assume we are doing focused attention meditation on the breath.

If I am remaining focused and not wandering, having any kind of feedback is likely to interfere with my focus.  If I am truly focused on the breath, I should be aware that I am focused on the breath.  I really don't need positive reinforcement.  On the other hand, if my mind is wandering and not focused, I may need to be alerted to this fact, reminding me that I should resume attention to the breath.  Over time, I may become more sensitive to the wandering and more able to readily return to the breath without the warning.

Auditory feedback is a readily available form of feedback, especially when we are already listening to instructions from the app via headphones or speakers.   However, it is common sense that attention to external stimuli such as sound will conflict with attention to the breath.  Whereas attention to the breath activates the interoceptive cortex (Farb et al, 2013), sound activates the auditory cortex.  If my mind is wandering anyway, this auditory "distraction" should not pose a great difficulty.  However, it should be momentary and only repeated long enough for me to recognize that I must redirect my attention.  If the auditory feedback is anything more than a quick reminder, it is likely to become a problem.

The Calm app provides different forms of auditory feedback.  Both the wind and the waves are giving feedback about the degree to which the mind is active.  The wind supposedly represents current activity, whereas the waves represent activity over a more extended time-frame.  Both the wind and waves vary in intensity and are continuous rather than discrete sounds.  The birds are supposed to represent a calm mind, and they reward you for maintaining calm over a period of time.  If I hear any of these sounds, they are very likely to become items of interest in themselves, thereby distracting me from focusing on the breath.  Also this kind of feedback is likely to stimulate evaluation and judgement--"I am doing well," "I am not doing well," and so on.

Whether different forms of feedback would make a difference or not remains to be seen.  Haptic feedback (see might cause less conflict than auditory feedback, for instance.  On the other hand, with auditory feedback it might be possible to provide more differentiated feedback that sensitizes the user to the type of activity that is going on.  In which case the feedback would not so much function as a warning (wind and waves) or as positive reinforcement (birds), but as information about what is going on.

These issues challenge the concept of using a device such as this as a meditation aide.  It is to be hoped that appropriate research will be done to determine whether the Muse (and other devices like it) can be useful for the purpose of assisting meditation.

Farb, N. A., Segal, Z. V., & Anderson, A. K. (2013). Attentional modulation of primary interoceptive and exteroceptive cortices. Cerebral cortex, 23(1), 114-126.

1 comment:

  1. I think people's reactions may depend on how they engage the process of meditation. In my case, audio per se doesn't tend to interfere with meditation, and indeed can facilitate it noticeably. For instance, I have had some positive experience with brain entrainment audio/music playing during meditation. My typical meditation tends to be focused attention mantra meditation, which involves sound so maybe there is less room for conflict.

    I have a Muse EEG as well, and I have noticed some conflicts arising around the feedback element. While it still tends to be helpful overall, and if I'm already relatively calm the feedback tends to be enhancing rather than distracting, there are definitely situations where there is conflict. For example, there are times when I'm just calm enough to get rewarded with the birds chirping, but then I notice that, and immediately the birds flutter away. That fluttering noise is probably the most distracting experience that I have had with the Calm app. Another example is sometimes if I adjust my eye position just slightly, I suddenly hear a huge blast of wind -- that too feels disproportionately distracting. [ Perhaps the latter may be harder to mitigate due to the large EOG signal generated by eye movements ].

    From what I've read, the error negativity signal in the anterior cingulate cortex may be involved. Error negativity during meditation can be particularly distracting, because at the very moment when I want to stabilize a level of calmness that I may have just barely reached, I feel rudely shoved out of it. Using the brain's feedback/reward loop to facilitate the process of letting go may indeed be a double-edged sword which needs to be wielded very delicately :-) I guess it's also possible that certain people may be more sensitive at certain times to this kind of negative feedback distraction than others.

    Again, if I'm already at a certain level of calmness, I can notice that the birds are chirping without being distracted sufficiently to cause the flutter. But that seems to be more a function of how calm I was before I started the session. What I'm looking forward to, from Calm and other neurofeedback applications, is an increased capability to help stabilize a level of calmness. Perhaps entrainment can be used to supplement feedback -- e.g. at critical moments, negative feedback is turned off and replaced by audio that actively promotes the calmer brain states.