OUR POSTHUMAN LEARNERS
AI AND THE FUTURE OF EDUCATION

Ryan Straight, Ph.D

College of Applied Science and Technology
University of Arizona

 

OUR AGENDA

The Plan for Today

1. Introduction
2. Terminology
3. Mediation (Postphenomenology)
4. Methods (Posthuman Inquiry)
5. Technology
6. Future

Link to this slide deck!

The plan for today is to explore the basic concepts needed in order to really dig into how these technologies interact with and afford changes in not just education as a big, nebulous concept, but in our day-to-day. As we’re still very much in the early stages of this, much of it will be necessarily theoretical or look toward potential rather than a postmortem. That said, we’ll do our best to tease these things out and at the very least provide a roadmap.

Next we’ll be laying the groundwork for the ability to use that roadmap. We’ll start with the foundations of the various ways in which technology mediates our experience of the world. This is, in part, to give us a framework and language with which to understand new technologies like AI. This is known as postphenomenology.

After that, we’ll take that groundwork and look at methods of application. A kind of, “Okay, great, but how?” And, specifically, how in education? This is what we refer to as posthuman inquiry.

Next, we’ll look at what this approach tells us about the future of this space. How can we make it successful? What do we need to be wary of?

So, let’s get going!

INTRODUCTIONS

Ryan Straight, Ph.D
Honors/Assistant Professor in Cyber, Intel, & Information Operations
Director, MA{VR}X Lab
College of Applied Science and Technology
University of Arizona

• Me, what I teach, et cetera.
• Going to be introducing you to a lot today, so you’ll leave with more questions than you probably came in with and I think that’s usually the better result.

TERMINOLOGY

Let’s speak the same language

1. Posthuman
2. Posthuman Inquiry & Postphenomenology
3. Artificial Intelligence (AI)
4. Virtual Pedagogical Agents

Image via dream.ai.

We need to be on the same page when we’re talking about these rather new topics, so I want to give a quick rundown of what these things actually are. I’ll caveat it with: so far, with just a few exceptions, there’s no absolute consensus here.

• Posthuman: some think it means being a cyborg or living in The Matrix. Others see it as having the ability to alter one’s self just as easily as the material world around us. Brain-computer interfaces. Just about any episode of Black Mirror. But it’s not necessarily just like that. For our purposes, we’re focusing on understanding moving beyond the traditional human-technology dichotomy, which opens us up to actually understanding much, much more about how we now exist in the world.
• Posthuman Inquiry and Postphenomenology: these are technically separate approaches to researching in this space but in our brief time we’ll use the terms more or less interchangeably. They are both methodologies, ways to say “okay, so how do I go about understanding this new existence?” We’ll be coming back to this shortly.
• Artificial intelligence: our current iteration of AI is “generative,” meaning we have have so-called “large language models” that are used to generate responses to prompts. Anyone that’s used ChatGPT knows how this functions from a user-standpoint, if not perhaps the underlying technicalities.
• Virtual pedagogical agents: exactly what it sounds like and this is where things get really fun.

MEDIATION

This brings us to “technological mediation.” What does that even mean? Some examples may shed light:

The Original Four

Don Ihde (1990), in his attempt to reconcile the “classical” phenomenology that came from the likes of Edmund Husserl, Maurice Merleau-Ponty, and Martin Heidegger with our technologically immersed existence, proposed a new version of phenomenology, a post-phenomenological approach. This “post” prefix is more akin to postmodernism rather than posthumous or postmortem, in that it refers to simply the next logical step rather than something that necessarily comes after. It is worth mentioning here that when Ihde was exploring this new paradigm, it was the 1970s, so consider “technology” (in all its forms discussed earlier) now and half a century ago.

Regardless, through his ponderings he decided upon four different “technics,” as he called them, or ways technology mediate our experiences. As he says, “Instruments are the means by which unspoken things ‘speak’, and unseen things become ‘visible.’” They are relational and include intention, which we’ll come back to in a minute. So, briefly, what are they?

Embodiment

(human – technology) → world

People and technology together relate to the world.

• You see through a telescope.
• You talk through a phone.
• There is technologic transparency.

Image via GIPHY

Here, the person more or less incorporates the technology into their outward, physical, proprioreceptory experience. A classic example is a blind cane. The cane, in essence, becomes an extension of themselves and their perception. Likewise, glasses or a telescope.

Importantly, here, we get the concept of transparency (Ihde, 1990), which, then, can be understood as the degree to which a technology recedes into the background during use. The actual conscious manipulation or awareness of the tool drops and one’s intention seems to flow effortlessly through to the ultimate target of it. A classic example here is a video game controller. As the player becomes more familiar with the tool, the interstitial steps to remember which button results in which command become redundant, as simply the thought of jump leads to the sprite jumping with no conscious effort on the part of the player. This is closely related to “field composition” or “field of awareness” (Rosenberger & Verbeek, 2015, p. 23), in which one’s actual perception shrinks or narrows, such as no longer noticing what happens beyond the edges of the screen when viewing a movie or to the contents of a slide while using a microscope.

So already you’re probably thinking, “These concepts feel familiar.” And that’s really the point: this is all a way to take these lived experiences and develop a methodology that allows for the empirical study of our lifeworld with these technologies. And what more relevant technology can you think of right now that is fundamentally changing how we engage with the world? Yup!

Hermeneutic

human → (technology – world)

• You read off a speedometer.
• We interpret an x-ray.
• We assume the translation is accurate.
• There exists a gestalt.

Image via GIPHY

The hermeneutic relation, on the otherhand, involves translating or interpreting technology in order to understand the world. You essentially read through the artifact, such as a speedometer or a clock. Writing, itself, is a technology that falls within this category. (Kline, remember?)

This is also where we begin to understand the gestalt, which, in this context, refers to more or less the settling in of the dominant, typical use of the thing that may provide even more information than is present. Think of a radiologist reading an x-ray and bringing to bear training to that interpretation: they can see more than what is actually there.

Alterity

human → technology (world)

• Technology as other.
• We’re in its system, not ours.
• World withdraws; we focus on the technology.

Image via GIPHY

In the alterity relation, the technological artifact is treated as itself, what Ihde deemed the “quasi-other.” This ranges from a blender to an Amazon Echo to a fully functioning robot. We will come back to the alterity relation later, so keep in mind the bullet points here, especially number 2.

Background

human → (technology / world)

• Impacts our environment.
• Through this, us.
• Often don’t notice until it breaks.

Image via GIPHY

Finally, the background relation describes experiences like an air conditioner: this is technology that influences and impacts the world around you but you have little to no interaction with. It happens in the background, surprising enough. Very often, the presence of these technologies only come into view when they break. For those who remember their Heidegger, they become vorhanden, or present-at-hand, and it’s their broken state that defines them.

BUT WAIT THERE’S MORE

Later, other relations were beginning to be identified as technology changed and the distance and intention from and with us changed, as well. For example Verbeek (2008) describes a few new and necessary hybrid intentionalities:

The Next Generation

Fusion/Cyborg

( human / technology ) → world

Composite

human → ( technology → world )

Augmentation

( human – technology ) → world
                    ↘ ( technology – world )

Here we start getting into more complex relations that you can imagine being used in cutting-edge or future online/virtual learning environments. Hopefully you’re beginning to see where I’m going with all this.

• First, fusion or a cyborg intentionality. Here we’re talking about everything from IUDs to pacemakers, from implanted RFID chips to cochlear implants. In this case, there is no physical separation of the self and the technology. In fact, removing one of them from the equation breaks the entire thing down. A pacemaker without a user has no impact, and a user without a pacemaker is… well. You get the point. The person and the technology are literally fused insofar as experiencing the world is concerned.
• Composite intentionality: Human -> ( Technology -> World ), wherein a person’s intention is directed toward the technology, and that technology’s intention is thereby directed at the world. While a thermometer represents the temperature in the hermeneutic relation, A thermal camera, for example, which translates a world we cannot naturally perceive (infrared radiation) into something we can (a colorful video) in the composite relation.
• Augmented intentionality: considerably more complex, it includes a kind of feedback loop, wherein the human’s intention directs the technology to “experience” the world, the results of which are then fed back to the human and the cycle repeats itself. So, not only is the human experiencing the world through the technology, the human is also experiencing the technology’s experience of the world overlaid, and reacts to that.

The question, then, is how many of these different kind of technological relations can you identify when engaging with artificial intelligence?

EDUCATION

Bringing It All Together: Where Does Education Fit In?

So, while we can clearly see how, through a hermeneutic relation, a speedometer can mediate your experience of driving a car, for example, but what happens when the goal of your interaction is teaching or learning and what happens when we consider AI through this lens?

Doing Postphenomenology

… in education involves attending to the unique differences a particular technology makes to teaching practice, knowledge apprehension, and pedagogical meaning.

As Adams and Turville (2018) point out, “Doing postphenomenology in education involves attending to the unique differences a particular technology makes to teaching practice, knowledge apprehension, and pedagogical meaning” (p. 20). The question then becomes, in this new version of technologically-mediated learning, what are these AI-related paradigms of teaching practice, knowledge apprehension, and pedagogical meaning?

Crucially, doing postphenomenology has no actual, agreed-upon method (Rosenberger & Verbeek, 2015). There are a variety of suggested heuristics (Adams & Thompson, 2016), but no hard-and-fast checklist or guide. This is where we start getting into posthuman inquiry, which is itself closely related to an applied postphenomenology. Again, you are absolutely welcome to explore the resources and delve deeper into how and where these differ, and I encourage you to do so!

But How?

1. Composing Anecdotes through Self-Observation
2. Gathering Lived Experience Descriptions through Interviews
3. Composing Anecdotes through Observation of Others
4. Studying Breakdowns and the Eidetic Reduction

Image via GIPHY

Accordingly, here are some ideas of ways you, too, can begin attempting to exploring this space and, if you spend time educating or being educated, the crossroads as well, as described by Adams and Turville.

1. Composing Anecdotes through Self-Observation
   1. Approach your own experiences and lifeworld with “non-intrusive observation of oneself” … and the “invitational appeals of one’s equipmental or technology-textured surround.”
   2. This is harder than it sounds. “Explanations, opinions, judgments, or theoretical concepts must be pushed aside in favor of what was given in the moment.”
   3. The point of this is, again, to “open up” access to the experience the technology affords, not just your opinion or feeling about it.
2. Gathering Lived Experience Descriptions through Interviews
   1. Perhaps not the most easily accomplished approach, to be fair, especially for those that aren’t necessarily in an environment where that kind of research is encouraged or even permissible.
   2. That said, this may be a tremendously revealing process, especially when directed toward teachers and learners.
   3. This should not be confused with user experience testing, of course. When gathering these lived experience descriptions, while the design choices made by software developers may have an influential and mediating impact as discussed early, it is merely one step to overcome to get to the real experience.
   4. While it is true that the application of the methods we’re talking about may sound similar to user experience, it’s important to note the difference that postphenomenology is concerned with the practical lens, not simply “does this thing work?”
3. Composing Anecdotes through Observation of Others
   1. Contrasting the interview approach, simply observing others in the space can also lead to revelations without overt disturbance.
   2. It may, for example, “assist in pointing up aspects of everyday life that may otherwise be taken for granted by oneself and others.”
   3. However, it’s important to note that, since through the observation you may not access a truly sufficient and representative lived experience (you can imagine just how much you would miss by watching someone’s actions and not being privy to their motivations or goals).
   4. This is the go-to when dealing with students, of course. Watch and learn.
   5. To quote Adams and Turville, “it is ultimately not the researcher, but the students, who are best positioned to describe how the world of the classroom and the science experiment appeared to them, that is, the intelligibility of the event to each individual, what material aspects of the experiment showed up as meaningful to them, and so on. Here, a postphenomenology of practice necessarily returns, not only to the things themselves, but also to the students themselves.”
4. Finally, Studying Breakdowns and the Eidetic Reduction
   1. Recall Heidegger and his broken hammer. In this approach, we come after. That is, instead of trying to tease out the lived experiences in vivo, it is the breakdown of the moment that gives us access.
   2. For example, when exploring how to understand AI in pedagogical contexts, it is easier to understand how people and things interact when there is friction than when it goes smoothly.
   3. Likewise, for a completely non-technical and more Heideggerian example: when your pencil or chalk breaks mid-thought. What happens to you? What happens to your train of thought? What happens to the creation you were mid-process with?
   4. One of the goals in this method is to not necessarily find those non-dominant stabilities of something but to tease out, after the “break,” what is invariant? What about it is unique?

Approaches

1. Variational method or analysis
2. Variational cross-examination
3. Case study
4. Conversational analysis

Image via GIPHY

We can make this even more concrete but first I need to briefly describe what postphenomenologists term “stabilities” and how technology is “multistable.”

Take a Necker cube, for example: you might immediately and effortlessly see a three-dimensional hollow cube in a particular orientation. With a little bit of effort, you can perceptually change to see it in another orientation. This is a great demonstration of a technology’s multistability: “a given technology has different meanings to different people at different times” (Wellner, 2020).

1. Variational method or analysis: described as “the method of brainstorming stabilities of a multistable technology.” Remember, everything is also something else.
2. Variational cross-examination: first determining what a particular technology’s “dominant” stability is, we can move on to critically examining and identifying alternatives by exploring, for example, habits of users or frequent qualities teased out of the technology itself, its role in a particular system (thinking back to it being just an actor in a network of actors), and what we are probably most comfortable with or at least what we tend to do with tools in education: what’s known as “material tailoring,” or “the … alterations of technology for different purposes.” Just think about how often we approach things in this space that way: “this has potential; how can we use it? How can this help me reach those learning objectives or outcomes?”
3. Third, case study: generally speaking the postphenomenological approach is this. If you wanted to generate a deep, meaningful understanding of how students use AI for a particular use case, this is how you would generally approach it. The first two steps are how you find what to study, though, again, it’s more a web than a straight line.
4. Finally, conversational analysis. This can be a little heady, but it can help us bridge the gap between simply exploring “what things do” and those mediations of our perceptions and actions “from within.”

I want to provide a concrete example of how this can be applied to real-world situations with a study by Robert Rosenberger on park benches. In that case, Rosenberger was looking at what’s known as “hostile design,” a way of designing things that make it difficult or impossible for one stability to exist. For park benches, it was the bars that separated a single bench into three individual spots for sitting. So we have a bench-as-bed stability now impossible, while the bench-as-sitting-location is maintained. The variational analysis of the bench then leads to this uncovering of multiple stabilities and thus leads to the uncovering of political choices being made in city design that target one group over another. Further analysis of local benches in a study by Galit Wellner found that, in her neighborhood, people would leave books or secondhand clothes on benches, revealing the bench-as-library stability and bench-as-wardrobe stability. This could further lead to deeper investigations of local neighborhood needs. Really neat method, I think.

So, again, as you can see, there is no hard and fast, prescriptive method to take. This, perhaps surprisingly, works in our favor. It requires the educator and researcher to delve deep to fully understand something that they likely never will, only the individual instance of the thing. In doing so, however, incorporating the posthuman inquiry or applied postphenomenological approach, we can “interview” these objects–whether tangible devices, virtual worlds, or digital personas like AIs–and give them time to “speak.” We musn’t simply explore how, for example, students “use” these things; we must conversely look at how those things use them.

THE TECHNOLOGY

Now that we have some ways of actually doing these things, let’s try to bring it back to the technologies like Web3, XR, and artificial intelligence before we triangulate with education.

Web3 & XR

Image via dream.ai.

I want to touch to Web3 and XR quickly before moving on to artificial intelligence because I think they’ll both be providing a supporting role going forward. Let me explain.

Web3 is a catch-all term for, in short, a decentralized, blockchain-based internet. If you got a chance to check out Dagan’s talk from yesterday, he did a great job explaining it, so I won’t spend much time on it here. So, I’ll just reiterate that it’s meant to provide ownership and agency while increasing transparency, trust, and accessibility. My best advice when it comes to thinking about education and Web3 together is to divorce any thoughts of cryptocurrency from it and look solely and the utility of the technology. Okay, moving on.

XR, or extended reality, on the otherhand, has a much more intimate relationship with AI in the realm of pedagogy.(And this example is not intended to be abilist; extended reality is visual-heavy medium, however.) I’d like to paint a picture for you:

You’re looking at a student. They’re sitting in a chair, no desk in front of them, with their hands raised, moving as if directing an orchestra. Their eyes seem to be darting randomly and they repeatedly angle their head slightly in various directions. What do you imagine is going on?

In probably less time than you think, we won’t need bulky headsets to engage with that augmented layer of reality that we talked about earlier. Instead, simply contacts will be enough. Perhaps a couple rings to make hand tracking easier.

There’s still the question of what the learner is experiencing, though. Perhaps they’re manipulating a photo-realistic human heart. Or a relief valve in a nuclear reactor. Or unearthing an artifact in the Saqqara necropolis. All of these are possible but, as it stands now, they all must be pre-developed. Somewhere within the process, someone had to sit down and code this experience into life. Think about all the curriculum you’ve designed, after all, just on paper. If there’s narration or interactive instruction involved, doubly so. But what if that wasn’t the case?

Artifical Intelligence

“Society cannot yet be programmed.”

Arnold Gehlen, 1983, A Philosophical-Anthropological Perspective on Technology

Image via dream.ai.

Pretty spot-on that Gehlen stated this exactly 30 years ago when we’re not starting to wonder, can it?

There is a clear sense in the educational community right now of both fear and excitement with the burgeoning AI landscape. So let’s unpack those a little bit:

There is genuine fear that AI will make students lazy, for lack of a better term. Perhaps, perhaps not. If a student is being asked to regurgitate facts, they may well go to ChatGPT because a) LLMs can be pretty good at it (and will soon be good enough that we can actually trust them), and b) with an almost Connectivist approach, the ability to access the information is often more important than the information, itself. Or, as Siemens said, “The pipe is more important than the contents of the pipe.”

Ask yourself: if I’m worried my student will use ChatGPT to do the assigned homework, where exactly is the friction? Is it with the nature of the assignment? Is it with my own fear of losing control? Is it evaluation that’s designed to simply test rote memorization? Is it simply the fact that the student lacks the intrinsic motivation to do what they might see as lacking value?

If we think of AI in terms of how Peter Paul Verbeek defines technologies, which “help to shape human actions, interpretations, and decisions that would have been different [otherwise],” we do really need to start thinking about the learner-AI hybrid entity as the student. Just as Frida Khalo needed a paintbrush to be Frida-the-painter and Florence Nightingale needed a lamp to be Florence-the-nurse, and Frida-as-painter and Florence-as-nurse are what allowed them to truly engage in the world in their most meaningful and impactful way, perhaps student-with-AI is what’s needed now.

Remember, AI is not smart, at least in its current form. “Artificial Intelligence is neither artificial nor intelligent.” AI is just very confident but it’s only as good as the input, right?

THE FUTURE

The subtitle of this talk is “AI and the Future of Education,” but it might be more accurate to think of “Education and the Future of AI.”

Virtual Pedagogical Agents

Currently

Soon

Image via dream.ai.

This is where we are now. But:

• The current majority of traditionally paced students grew up with smartphones (I’m avoiding using terms like Gen Z as they’re just very poor categorizations and come with a LOT of baggage). A window to the world at all times.
• Over the next decades, students in this same traditional pace will grow up with their own AIs. Like a valet. (Pitt, 2011) “Someone” they grow with and that can push them constantly just outside their comfort zones while keeping the instructional approach and materials appropriate. Think of The Diamond Age with Nell and Miranda but without a live “ractor” on the other side.

However, a question arises: well, a few, actually, that I invite folks to think about when we consider ways to both introduce AI to students and understand how it might impact them on an individualized level. I and a student just had a paper published in the EdMedia + Innovate Learning proceedings that relates to this, actually. (Straight & Yowika, 2023)

But, even if we’re not at the AI place for this yet, it’s worth considering these questions for instructional design, alone.

Preparing Students

1. Support students and staff literacy
2. Staff well-equipped
3. Ethics first
4. Academic integrity
5. Best practices

Image via dream.ai.

As if students need us to prepare them for using AI. Ethically using it in a traditional educational setting, perhaps, yes. So, let’s look at what one group has suggested institutions do regarding the adoption of AI and ensuring proper student preparation. These five guidelines are generally good but a tad myopic, I fear. Still, a great starting point. The reason I feel this way has more to do with the timing. This is not just a new tool that students will be more familiar with than staff; it is a new way of experiencing the world and understanding the nature of knowledge, itself (which, hopefully, I’ve sufficiently laid out earlier).

To quote Andrew Brass, head of the School of Health Sciences at the University of Manchester

We know that students are already utilising this technology, so the question for us as educators is how do you best prepare them for this, and what are the skills they need to have to know how to engage with generative AI sensibly?

From our perspective, it’s clear that this can’t be imposed from the top down, but by working really closely with our students to co-create the guidance we provide. If there are restrictions for example, it’s crucial that it’s clearly explained to students why they are in place, or we will find that people find a way around it.

I keep getting reminded of the phrase: “build a better mousetrap and nature builds a better mouse.”

But again, we’re still seeing approaches as generally reactive to the current state of AI, so is this enough? We’re in for probably one of the most difficult but exciting times since education has been a profession.

I think the most important thing here is ensuring that students are intrinsically motivated. So how does one foster intrinsic motivation in learners who are more?

Our Posthuman Learners(?)

Image via dream.ai.

So, are our learners posthuman? If not yet, will they be?

If we believe that, yes, that certainly is or will be the case, we need these new ways of understanding the exploration and examination of the learner/technology hybrid that aren’t mired in traditional, perhaps out-moded approaches.

And, not to put too fine a point on it, but we’re hopefully in a position to realize that approaching AI and these other technologies going forward requires more than just some bullet points of action items: it becomes a philosophical discussion, an ethical discussion. Just think about the virtual pedagogical agent and compare a student’s lifelong AI with, for example, Deckard and Rachael in Blade Runner: should an AI persona be treated as human? What are the potential ethical consequences of learners treating an AI as a teacher, as a collaborator, or even as a servant? AIs that are anthropomorphized are drawing criticism, with calls for AIs to not use first-person singular language. “This language model is not built for…” compared to “I am not able to….”

For example, we know that when dealing with virtual pedagogical agents, those agents are important because they have a number of desired traits (or can, at least): adaptive, versatility, realistic, match students socio-culturally, foster commitment, motivation, responsibility, and improve learning (Alfaro et al., 2020). There is evidence for the traits of a virtual pedagogical agent having distinct and key influence over a learner. Gender-same agents have been shown to be effective in bringing girls into technical areas (often seen as stereotypically male, like cybersecurity or computer science) and help fight that stereotype. Conversely, female-presenting agents tend to be trusted and listened to more in assistant, social influence, and cosmetic-based roles. Male-presenting agents are trusted more in areas like sports. The classic “professor” look (you can guess what that is) elicit more feelings of expertise. And interestingly, even when an agent has no prescribed gender presentation, learners tend to ascribe a binary gender to it. This could have tremendous importance going forward when we, again, begin seeing students with their own personal AI that has (in all probability) helped raise them.

Either way, again, we absolutely need to start approaching our learners not simply as a student using a tool, but as the learner-AI hybrid entity and teach to that.

Back to the Dancing Doggo

 

Now back to the dancing doggo from the beginning and why it’s relevant for a talk on the future of AI and education.

Anyone want to hazard a guess as to what that music is on the right?

When Chopin saw a little dog dancing, he was inspired to write the Minute Waltz. It was the product of billions of neurological connections–some trained, some coincidental but serendipitous–that resulted in the music. It was not a well-engineered prompt. These connections are what drive creativity and what education feeds on. LLMs are external to us but they are a reflection of our own thoughts and ideas, manifested as data accessible to be scraped on the internet. LLMs can mimic creativity but not actually be creative. I’d suggest that AI may lead to a burst of creative effort but not actually creative production, in that prompt engineering is not an analog for the emotive expression of creating with paint and a brush, for example.

Will this continue to be the case? Your guess is as good as mine. What is for certain is, I think, that our learners are heading inexorably toward being posthuman if they’re not there already. The best thing we as educators can do is ride the wave and stop trying to out-smart the new landscape.

THE END

That’s a wrap!

Thank you for coming! Find out more at:

MA{VR}X Lab

https://mavrxlab.org

Me

https://ryanstraight.com

Link to this talk!

https://mavrxlab.org/news/2023-07-16-kti-c/

Thank you, everybody, for taking a chance on a session like this. I know it’s not the usual but I hope it’s interesting and novel enough that it’s sparking some creativity and curiosity. You can find my contact information on the KTI-C website, here in the slide deck, and on my lab’s website. On the lab’s website you’ll also find direct links to the Pennsylvania standards, which I’d encourage folks to look at through the lens I presented today. Maybe that spark will appear!

Now, happy to take any questions or comments or thoughts anybody has!

References

Adams, C., & Thompson, T. L. (2016). Researching a Posthuman World. Palgrave Macmillan UK. https://doi.org/10.1057/978-1-137-57162-5

Adams, C., & Turville, J. (2018). Doing Postphenomenology in Education. In Postphenomenological Methodologies: New Ways in Mediating Techno-Human Relationships (pp. 3–25). Lexington Books.

Alfaro, L., Rivera, C., Luna-Urquizo, J., Castañeda, E., Zuñiga-Cueva, J., & Rivera-Chavez, M. (2020). New Trends in Pedagogical Agents in Education. 2020 International Conference on Computational Science and Computational Intelligence (CSCI), 923–928. https://doi.org/10.1109/CSCI51800.2020.00172

Ihde, D. (1990). Technology and the Lifeworld: From Garden to Earth. Indiana University Press.

Pitt, J. C. (2011). Doing Philosophy of Technology (Vol. 3). Springer Netherlands. https://doi.org/10.1007/978-94-007-0820-4

Rosenberger, R., & Verbeek, P.-P. (2015). A Field Guide to Postphenomenology. In Postphenomenological Investigations: Essays on Human-Technology Relations (pp. 9–41). Lexington Books. https://research.utwente.nl/en/publications/a-field-guide-to-postphenomenology

Straight, R., & Yowika, W. (2023). From Parasocial to Posthuman: (Virtual) Pedagogical Agents, Parasocial Phenomena, and the Future of Immersive Learning. Proceedings of EdMedia + Innovate Learning, 1028–1033. https://www.learntechlib.org/primary/p/222614/

Verbeek, P. P. (2008). Cyborg intentionality: Rethinking the phenomenology of human-technology relations. Phenomenology and the Cognitive Sciences, 7(3), 387–395. https://doi.org/10.1007/s11097-008-9099-x

Wellner, G. (2020). The Multiplicity of Multistabilities: Turning Multistability into a Multistable Concept. In G. Miller & A. Shew (Eds.), Reimagining Philosophy and Technology, Reinventing Ihde (pp. 105–122). Springer International Publishing. https://doi.org/10.1007/978-3-030-35967-6_7