Way back in June, I planned to write a post prompted by Kevin Kelly’s talk at the Quantified Self conference in May and a new word I’d heard in an interview with David Brin. Between then and now, the summer months have whipped by, so please excuse the backtracking – those of you who have seen the site before will have heard of our bodyblogger project, where physiological data is collected on a continuous basis and shared with others via social media sites or directly on the internet. For instance, most of the time, the colour scheme for this website responds to heart rate changes of one of our bodybloggers (green = normal, yellow = higher than normal, red = much higher than normal – see this for full details). This colour scheme can be mapped over several days, weeks and months to create a colour chart representation of heart rate data – the one at the top of this post shows a month’s worth of data (white spaces = missing data).
In this context, I was very interested by Kevin Kelly’s concept of Lifestreams as presented at the QS conference.
“What emerges in this new model are Lifestreams. That’s what we curate in the age of the quantified self. We head upstream, and we leave a wake of data behind us. Lifeloggers, who log everything they do, are pioneers in this space… take these exercises to an extreme, and they’re sharing it, as part of the shift from me to we.”
This quote is taken from Ethan Zuckerman’s liveblog of Kelly’s talk, you can read his full transcript here.
The bodyblogger project is a very potent demonstration of a Lifestream in action. It’s continuously available, always changing and open to interested observers. To date, we have focused on different use cases for the bodyblogger system, such as: out-patients communicating health status to medical staff or family members, sharing physiological experiences such as fitness programmes etc. However, as Kevin Kelly points out as we are fast approaching a period where monitoring technology is mobile and pervasive, so there will never be a moment when we are not logging activity or interacting with devices that can monitor activity. Instead of looking for use cases to somehow justify the need for pervasive monitoring, we should perhaps think about collecting and sharing physiological data for its own sake and let the use cases develop from that in a bottom-up fashion. Back to Kevin Kelly’s talk, quotes from the liveblog:
“These lifestreams intersect with each other and are, in a way, creating a new media. If we organize computation around lifestreams, an intersection between our lifestreams is a communication, an event of some sort. The media we are in is these streams of data… This environment, with data streams and life streams, is the space where we’ll do the work of the quantified self.”
This first sentence seems especially prescient in my opinion. Bodyblogging and related data (GPS coordinates, physical activity sensors) are a new form of media; like existing media forms, the main purpose is to communicate, unlike existing forms of media, it has the potential to communicate as much information to the individual bodyblogger or lifestreamer as to an audience of interested spectators. Allow me to elaborate -if the biofeedback paradigm teaches us anything, it is that many physiological processes are not directly available to conscious inspection and making those processes open to inspection changes the relationship between the person and his or her body. This earlier post from Ute reflected on this kind of mechanic based on her experiences with moodscope and bodyblogging; I also wrote this post about the potential downside of this development. If you are inside this media or stream of data, your level of self-awareness is enhanced which is one route to augmented self-regulation, whether it be anger management, dieting or reducing your experience of stress.
If your data stream is opened up to other technological systems (your phone, your electronic diary, your car, your house), there is enormous potential for implicit monitoring to inform a plethora of so-called ‘smart’ technology, as they adapt themselves to this data stream. Your mobile could switch to silent mode if you’re deep in conversation, your stream of emails could be delayed for five minutes if you are highly engaged with document editing, your music player could select relaxing music if you seem particularly agitated. These systems represent an intersection between bodyblogging and biocybernetic adaptation (i.e. implicit adaptation of software in response to physiology) – in fact, it posits that biocybernetic adaptation will become the dominant paradigm in HCI provided that sensors and associated software become pervasive in modern life.
We’re currently working on a project to detect interest levels via psychophysiology. The rationale for this work is based around Herbert Simon’s idea that the “cost” of information provision is attention. Given that selective attention is finite but the capacity for technology to present information is almost infinite, it stands to reason that a pre-conscious filter is necessary to shield us from multiple distractions and the threat of perpetual information overload. If we can achieve that function via a bodyblogging system that monitors in order to learn about preferences, and on that basis, to edit your exposure to information on a pre-emptive basis, this is a worthwhile endeavour. In addition, it would yield a record of this process via the bodyblogging data stream that the user can peruse to their heart’s content.
It is with some reluctance that I wish to draw an analogy between the common garden snail and adopters of this nascent technology (especially as several are good colleagues), but I think we can compare bodyblogs and lifestreams to the trail of slime left by a snail as it passes through the world. The slime secreted by the snail acts as a record of the journey, it is also a means of transportation for the animal. Data streams of the physiological day also function as a record of the day, quantifying our physical movements as well as fluctuations in psychological states. In a future world, where information technology is constantly crying out for attention, it may also ease our path through the information landscape of virtual distractors without losing mental focus. Given that the original human-tech hybrid, the cyborg, was designed to live in the hostile environmental conditions of space – one wonders if we will require technological augmentation of this kind simply to function in the digital space of the future.
As a second point, bodyblogging and lifestreams are extrapolated in terms of sharing information to others or as Kevin Kelly says facilitating the shift from me into we. But of course, it doesn’t have to be that way. Bodybloggers may be motivated to quantify the self as a form of private self-knowledge, not to be shared with other people. If you wanted to test whether a family member or even a spouse caused your stress levels to increase, you may want to collect that data privately. Similarly, if bodybloggers are self-monitoring to deal with personal health problems, like incontinence or erectile dysfunction, they would probably not want to post that data to a Facebook page nor would many thank them for doing so. The point is that bodyblogging doesn’t have to be about sharing data in a public or semi-private domain, the concepts works as a private activity. But if we do decide to share our data, how does that work? Kevin Kelly envisions data functioning as an overlay – again this quote is from liveblog of his QS talk:
“What we’ll see very soon is spectacles and glasses that will let us see that data world. It might be a screen we hold up, but we’ll be able to see this overlay of the digital world embedded in the material world. There’s thinking that the digital life is disembodied bits… but it’s really about bits embedded in the physical world.”
With reference to bodyblogging and physiological datastreams, we’re talking about digital embodiment as an overlay on physical reality. There are two levels to this type of data depending upon whether we draw physical or psychological inferences from the physiological data. With respect to location and physical movement, there is a ‘bread-crumb’ model here where people could elect to anonymously dump data as they pass through a physical location. For instance, I could wear my digital glasses and see an accumulated record of every person’s heart rate who had walked up my street or the steps up to my office. There is no reason why I would want to do that as an individual, but for architects or others with an interest in planning public spaces, they may find a use for these data. Similarly, with an eye on research into intelligent cities, it would be interesting to obtain dynamic data patterns from groups of people in order to adapt technology on the basis of populations rather than individuals. As an aside, here is a link to a slideshow from Matt Jones at BERG about people as walking architectures, which touches on this kind of concept, but not for physiological data as such.
A potential strength of representing bodyblogging data as a crowd-sourced aggregation is that this method protects the identity of the individual, provided that the way in which it has been aggregated cannot be reverse-engineered. In addition, like the slime-trails of snails, these data streams can have limited availability and disappear over a period of time, further protecting the privacy of the individual bodyblogger.
The notion of sharing physiological data always raises the issue of privacy. I read David Brin’s book “The Transparent Society” some time ago and enjoyed his take on surveillance and privacy, so I was interested by the interview with him here where he talks about the notion of sousveillance, where everyone (citizens, government) is capable of monitoring everyone else. His primary motivation in putting this idea forward is to counteract the use of surveillance technology by government agencies (and now, at least in the UK, journalists) by making the same technology available to ordinary citizens – so in principle, we could eavesdrop on the eavesdroppers. This seems to me to be an exercise in damage limitation rather than a genuine need, i.e. the more citizens know about government, the less potential for government to abuse their powers. With respect to bodyblogging, if we have a scenario where data is being shared in public spaces, we have a sousveillance situation but with the strong positive that nobody can be disadvantaged by the availability of these data if it is impossible to trace it back to the individual. We use a similar system in many psychology experiments where participants are identified by a number that cannot be traced to any record of their name or contact details. The primary directive (it seems to me at least) for data sharing should be that the data cannot be used to exploit an individual and of course that the individual maintains full control and consent over the parameters of data sharing (i.e. sharing is not the default position, it must be selected by the user who gives full consent).
Now, how about the second way in bodyblogging can be shared where we are able to infer psychological meaning from the signals? Here we enter a strange, new world in this case, where digital content equates to a record of our inner life. For many of us, our inner life of thoughts or feelings is something closely guarded and only shared with friends and family members. There may be occasions when we can share these data in order to enhance a shared experience, e.g. white-water canoeing, scary fairground rides, bungee-jumping, networked computer games. These data opens up a parallel world of digital feelings and digital memories, which may diverge in some respects from our literal experiences and recollections. Again, we can hide our data into crowd-sourced databases in order to protect identities or encrypt the data for sharing with a select few. The problem is that interpretation of the signals is bound up with our own identities and the context of our relationships with others. The concept of sousveillance doesn’t provide a strong fit with this case, the potential gain of sharing these data does not seem to be there. It also seems to me that bodyblogging for psychological insight is not just a private activity, it is also a retrospective one as exemplified by the affective diary project. One can only imagine the disruptive effect of conducting a conversation with a person whilst viewing a real-time feed of their physiological responses to your own words – fun in a gameplay scenario, not so great if you were asking someone out on a date.
Lifestreaming and bodyblogging are a new form of media that can represent private or shared experiences, the use of this technology has the potential to be beneficial for individuals and groups, regardless of whether users use the resulting data to focus on me or we.
