Exploring the Phase-Space of Information Architecture

Part 1. Introducing the phase-space of information architecture

Information architecture (IA), as a community of practice, worries most about meaning, where meaning has dependencies on structure. The concept of “structure” has recently expanded from single artifacts (e.g., a website viewed on a desktop computer) to cross-channel systems (HTML morphing across the multiverse of back-lit devices, data streams from the Internet of Things) (Resmini & Rosati 2011; Morville 2011; Fitzgerald 2014). IA practitioners appreciate that the stuff of meaning comes in different flavors: perceptual information (pre-attentive, physiological) and linguistic information (concepts and terms established by convention), where the qualities of meaning are subject to ever-changing mixtures of goals and context(1). Each flavor of information—perceptual and linguistic—is multi-faceted. These facets, taken as raw materials of design, effectively act as dials information architects may use to calibrate semantic interaction potential and tune information structures to facilitate what practitioner-theorists have dubbed, “places made of information” (Arango, Hinton, & Resmini 2011; Covert 2013; Klyn 2014; see also Dourish 2004 for “place” from the HCI point of view).

If we turn the dials and run all the permutations on the facets of our raw materials of perceptual and linguistic information, we generate what I will call the phase-space of information architecture. To visualize this, consider water. The nature of water depends on the combination of pressure and temperature at any given time. If we turn the dials of pressure and temperature through all of their permutations, we end up with the phase-space of water (figure 1).

Figure 1. The phase-space of water.

Now we may point to temperature/pressure combination neighborhoods where water lives as solid or, over at extreme temperatures, as plasma. We see the boundaries where, with a nudge either way, the phase shifts from solid to liquid, liquid to solid. We call out these boundaries and points of confluence and inflection as features, and we name them. This is how we understand the contextual dynamics of what it means to be water. The question I ask here is this: is information like water? Does it have its own state phases when we consider varying combinations of perceptual and linguistic information? What are these phases and what should we call them? What would a mapping of the entire phase-space look like? Most importantly: what would it tell us about our raw materials of design?

Figure 2 imagines what a phase-space of information architecture might look like.

Figure 2. Imagining the phase-space of information architecture. [The dragon in the figure is a Public Domain, no copyright image by James Keuning, via The Noun Project.

There would be shared regions where the flavors of information vary in dominance, labeled here, think our way (linguistic dominance) and sense our way (perceptual dominance) We’d find a pre-attentive domain where linguistic information is absent or quite low. We may try to imagine areas of linguistic-only information, but could we really ever walk away from all perceptual information and achieve a kind of semantic plasma with pure language? Sensory understanding of our position in space/time persists and our stomach may rumble while we are supposedly melding with the pure association world of ideas. It’s fair to expect we would find a semantic Forbidden Zone at the lowest fringes of perceptual information.

Maybe the interactions of perceptual and linguistic information are not fully captured in two dimensions. In the neighborhood where perceptual information is intense, affective responses (emotions) are evoked, which in turn may engage conceptual attention. We would likely need a third phase-space dimension to note how information morphs on the continuum of familiarity, from novel to tacit. Trajectory is important: frequently used linguistic information in the same visual (perhaps even auditory) form is so familiar it becomes like perceptual information that no longer needs active attention (the word “stop” on a traffic sign, labels on familiar software icons). In the context of designing for dynamical systems, we may want to map trajectories around the phase-space afforded by our structures. Sometimes the information included in a dynamical system would all take place within a single neighborhood; other times it would be a shimmering: the sense our  way neighborhood lights up with activity for a bit, then a stretch of pre-attentive reflex interaction, then some blips up in think our way, etc. Imagine plotting findings from contextual inquiry research in the phase-space diagram. Imagine creating two diagrams for a given project: an as-is, and a to-be mapping of the potential phase-space trajectories to show a desired offloading (or increase) in active attention, as the design goals suggest. An IA phase-space diagram could be a personal thing, idiosyncratic to each designer/design problem, or could be the product of discussions within the IA community (informed by cognitive psychology), with collaboration on phase boundaries and what we should call them.

We don’t need to know exactly what the phase-space diagram would look like to detail the facets, or phase-space dials, of the two flavors of information: perceptual and linguistic. We will do that next, in part 2 of this series. Part 3 will take us through a design example that uses an extreme region of the information architecture phase-space, the place where we turn the perceptual dial all the way up, the linguistic dial all the way down (or as far we’re able, when the time is right), for a very specific situation: the exploration of abstract concepts in a complex domain for the purpose of information discovery and education by way of a phase-space tuned, embodied information architecture. Closing the series, I will hint that this seeming extreme example may not be that different in nature from what we face as information architects designing structure for dynamical systems.

Continue to Part 2.


1. Investigating just how tightly coupled meaning is to shifting affordances of the environment, where language is just another kind of environmental information, is a growing thread among ecological psychologists investigating embodied cognition (Wilson & Golonka 2013; Chemero 2011). Embodied cognition is interesting to information architects because it matches in kind the things we worry about: contextual affordances and meaning through goal-directed action (Hinton 2013). Dourish has also adopted the term “embodied interaction” for HCI (2004).


Arango, J., Hinton, A., & Resmini, A (2011). “More than a metaphor: making places with information.” Information Architecture Summit 2011. Hyatt Regency, Denver, CO. Retrieved 20 February 2014 from http://library.iasummit.org/podcasts/more-than-a-metaphor-making-places-with-information/.

Chemero, A. (2009). Radical embodied cognitive science. The MIT Press.

Covert, A. (2013). “Making sense of place.” Midwest UX 2013. UICA, Grand Rapids, MI. [lecture] Retrieved from, http://www.slideshare.net/AbbyCovert/making-sense-of-place-midwestux-2013-keynote.

Dourish, P. (2004). Where the action is: the foundations of embodied interaction. The MIT Press.

Fitzgerald, A. (2014).  Architecting the connected world. O’Reilly Radar. 21 February 2014. Retrieved 21 February 2014, from http://radar.oreilly.com/2014/02/architecting-the-connected-world.html.

Hinton, A. (2013). The world is the screen: elements of information environments. Information Architecture Summit 2011. Hyatt Regency, Denver, CO. Retrieved 20 February 2014 from http://www.slideshare.net/andrewhinton/the-world-is-the-screen.

Klyn, D. (2014). Is it usable yet? IxDA Portland. Mozilla, Portland, OR. [lecture] Retrieved 20 February 2014, from http://understandinggroup.com/2014/01/ixdaportland/.

Morville, P. (2011). Editorial: the system of information architecture. Journal of information architecture 2:3, p. 1-8.

Resmini, A. & Rosati, L. (2011). Pervasive information architecture. Morgan Kaufmann.

Wilson, A. & Golonka, S. (2013). Embodied cognition is not what you think it is. Frontiers in psychology 4:58.


Marsha Haverty started her career in 1999 as an information architect, working with clients like Intel, DirecTV, Grainger, Reuters, Countrywide, Cisco, Microsoft, ASUS, and Autodesk. Currently at Tripwire, she designs data visualizations and contextual classification systems to help information security professionals understand the security posture and attack surface of their enterprises, and shapes UX strategy for future product direction. She has a long-held interest in exploring how information architecture may grow from community of practice to discipline.

Twitter: @mjane_h