3.1. Conditional Arising of Self-Understanding: Critical Perspectives on Technologically Mediated Environments
In Buddhist contemplative theory, self-understanding is not the discovery of fixed psychological traits residing within the individual. It is a contemplative process of observing conditionally arising phenomena. The essentialist view—which posits the self as an independent, invariant object—cannot adequately account for how self-awareness is formed and transformed in the actual field of practice, where sensation, emotion, and complex technological environments intersect in real time. Accordingly, the conditions under examination in this study are not treated merely as contextual factors that influence an already-constituted experience. They are understood as constitutive of the very structural possibilities through which self-understanding can become intelligible in the first place.
Although this study engages contemplative phenomenology, Madhyamaka philosophy, and philosophy of technology, it does not seek to synthesize them into a single theoretical framework. Rather, these traditions are employed as complementary analytical perspectives addressing different dimensions of the same phenomenon. Contemplative phenomenology helps illuminate the structure of lived experience, Madhyamaka philosophy provides a framework for understanding conditionality and the absence of inherent essence, and philosophy of technology assists in examining how technological systems participate in shaping the conditions under which self-understanding emerges.
This concern directly intersects with the McMindfulness critique that has gained momentum in recent religious and meditation studies. Scholars have shown that mindfulness has been reconstituted as an instrument of individual efficiency and psychological stabilization, recontextualized outside the ethical and soteriological frameworks emphasized in traditional Buddhism—specifically, the threefold training of sīla (戒, ethics), samādhi (定, meditation), and paññā (慧, wisdom) (
Purser 2019). This shift has unfolded across cultural settings as mindfulness is decontextualized and commodified (
Wilson 2014), and it belongs to the larger formation of Buddhist modernism—the synthesis of traditional Buddhist practice with Western psychology and scientific discourse that
McMahan (
2008) has examined in detail. That framework offers resources for interpreting AI-based meditation as a possible extension of Buddhist modernism. More critically, scholars have argued that alongside meditation’s interiorization, its interpretive horizons may be reorganized or diminished in ways that exceed mere cultural adaptation (
King 1999).
Such tendencies are likely to intensify in digital meditation applications and AI-based environments. Digital Religion scholarship has tracked how technological mediation reconstitutes the temporality, corporeality, and communal relationality of religious practice (
McDonnell 2014;
Geraci 2014). Research in this field demonstrates that digital technology transforms not only the content of religious practice but the structural conditions under which practice occurs—reshaping the form of religiosity itself (
McDonnell 2014). AI-mediated meditation may therefore represent more than a change in tools alone. It may reorganize the practical and interpretive conditions through which practitioners access self-understanding. The concern of the present analysis is not technological mediation itself, but how different configurations of mediation may shape the visibility of conditions and the location of interpretive authority.
This study extends these discussions by reinterpreting the secularization of meditation not as a shift in content but as a structural transformation in the conditional configuration that constitutes experience. In contemporary AI-based meditation environments, where algorithms instantly datafy and define practitioners’ states, self-understanding can no longer be treated as a purely inner operation. It must be grasped as the outcome of conditional arising—reshaped by technological mediation and environmental arrangement.
Grounded in dependent origination, this reconceptualization holds that self-understanding becomes visible only when sensory experience, attentional intentionality, and technological interfaces combine in specific ways. Rather than possessing a stable, invariant essence, the self is a fluid process that arises, transforms, and ceases as conditions shift—a position articulated in Buddhist philosophy under the concept of niḥsvabhāva (無自性, the absence of independent self-nature), which is examined in detail in Section Conceptual Framework: Conditional Structure Analysis Based on Dependent Origination. This non-essentialist view is what opens a reflective space for practitioners: by observing how conditions configure their experience, they can engage with that configuration actively rather than accepting technologically mediated interpretations as given.
Conceptual Framework: Conditional Structure Analysis Based on Dependent Origination
This section does not aim to provide a comprehensive doctrinal account of dependent origination. Rather, it reconstructs selected Buddhist concepts as analytical categories for examining how self-understanding becomes possible under different configurations of experiential and technological conditions. The concept of pre-configured self-understanding requires more precise specification than its three defining features alone provide. What, exactly, is configured in advance; at what stage of experience this configuration operates; and what structural consequences follow for agency and interpretation—these are analytically distinct questions, and the argument of this study depends on not conflating them. The following paragraphs address each dimension in turn.
At the level of what is configured, pre-configuration operates across three experiential strata. The first is attentional structure: the directionality and selectivity of attention—what the practitioner is guided to notice, and in what sequence—is organized by algorithmic interface design before phenomenological engagement begins. The second is interpretive framing: the categories through which sensations and emotional states are named and assigned meaning are supplied by the system prior to the practitioner’s own meaning-making. The third is perceptual salience: the relative prominence assigned to certain experiential elements over others is shaped by datafication and classification processes that precede conscious awareness.
A second dimension concerns the temporal position of pre-configuration: not what is organized, but when—that is, at what stage of experience this organization takes effect. Pre-configuration operates prior to the arising of experience itself, at the stage of condition-arrangement. This is what distinguishes it from scaffolding or guided attention, both of which intervene during or after experience.
A third dimension concerns what follows structurally from this prior arrangement, particularly in terms of how interpretive agency is distributed. At the level of structural consequences, the result is not the elimination of practitioner agency, but a potential redistribution of interpretive functions between practitioners and technological systems. In AI-mediated environments, classifications, feedback systems, and interface design may shape how experiential states are categorized, presented, and made available for reflection. The consequence is not the elimination of agency but its structural repositioning: the practitioner operates within interpretive possibilities that may already be partially structured in advance.
The analysis now shifts from the phenomenological register—how experience is structured and felt—to the ontological register of Madhyamaka philosophy, which provides the structural vocabulary for understanding why no element within that experiential field possesses fixed, independent essence. This is not a change of subject but a deepening of the same question: what makes pre-configured self-understanding structurally possible is precisely the absence of any essence that could resist reconfiguration.
Table 1 reconstitutes the principles of dependent origination as a qualitative analytical tool for tracing how self-understanding forms. In Buddhist philosophy, dependent origination (pratītyasamutpāda) is not a description of causal relations among discrete things. It is a structural claim within Madhyamaka philosophy: experiential phenomena arise through the interdependent configuration of multiple conditions (pratyaya), and no phenomenon possesses an independent self-nature (svabhāva). The formula in the Nidāna-saṃyutta (SN 12) of the Pāli Saṃyutta Nikāya—“when this exists, that comes to be (imasmiṃ sati idaṃ hoti); with the arising of this, that arises (imassuppādā idaṃ uppajjati)”—should be read not as a linear statement of causation but as a structural model in which experience arises within a configuration of conditions. This structural precedence, namely the arrangement of conditions prior to experience itself, is what grounds the concept of pre-configured self-understanding introduced in
Section 1.
Nāgārjuna’s Mūlamadhyamakakārikā (MMK) 24.18 deepens this insight. His declaration—“whatever is dependently co-arisen, that is explained to be emptiness (yaḥ pratītyasamutpādaḥ śūnyatāṃ tāṃ pracakṣmahe)”—presents śūnyatā (空) not as mere absence or nihilism but as the claim that all phenomena exist only in conditioned relationships, possessing no independent structure of arising. Self-understanding, accordingly, is not a substantial process that arises spontaneously from within; it is a relational process emerging through changing relations among sensation, attention, emotion, social context, and technological mediation. This is why this study emphasizes niḥsvabhāva (無自性): self-understanding is not the discovery of a fixed self but a non-essential, conditionally constituted process that can be pre-configured in particular directions.
This study reconstitutes these core concepts as analytical categories for tracing self-understanding formation. Rather than treating self-understanding as fixed self-recognition or the expression of inner dispositions, the framework grasps it as a process that arises, transforms, and ceases as conditions combine and change. The subsequent comparative analysis is structured around three axes: (1) what conditions constitute experience; (2) how those conditions are configured; and (3) how changes in that configuration transform the form of self-understanding. This redefines self-understanding not as the product of an autonomous subject but as a conditionally constituted process emerging through the arrangement and variation of conditions. One conceptual distinction warrants explicit attention. Anātman (無我) describes the absence of a fixed self in primarily psychological and contemplative terms—a concept native to early Buddhism. Niḥsvabhāva (無自性) articulates a more fundamental philosophical position: all beings lack an independent essence, and in Madhyamaka philosophy this is identified with śūnyatā (空性) (
Williams 2009). This study uses niḥsvabhāva to analyze self-understanding as a conditional construct rather than an independent entity.
Grasping (upādāna) occupies a pivotal place in the twelve-linked chain of dependent origination (dvādaśāṅga-pratītyasamutpāda), arising after craving (tṛṣṇā). It denotes the cognitive and emotional binding through which a particular interpretation of experience or state is identified with as if it were a fixed entity. In the present analysis, grasping is connected to the tendency of AI-based meditation practitioners to rely on algorithmically mediated interpretations without sufficient reflective examination.
This study draws on concepts from both Early Buddhism (Theravāda tradition) and Madhyamaka philosophy. However, this does not imply a doctrinal equivalence between the two traditions; rather, it reflects a deliberate analytical distinction between different levels of inquiry. In this framework, the Early Buddhist notion of anattā (non-self) is employed as a contemplative foundation for analyzing the experiential dissolution of the self within practice. By contrast, the Madhyamaka concept of niḥsvabhāva (lack of self-nature) is used as an analytical framework to interpret the absence of inherent essence in all phenomena. By assigning these concepts to distinct analytical roles—anattā for the analysis of experience and niḥsvabhāva for the structural interpretation of self-understanding—the study maintains a clear separation between contemplative and structural levels. This distinction enables a more precise account of how self-understanding is both enacted within practice and interpreted at the level of conditional structure, without collapsing the differences between the two traditions into a single doctrinal synthesis.
3.2. The Conditional Structure of Self-Understanding: Traditional Meditation and AI-Based Meditation
3.2.1. Delimiting the Scope of Analysis: AI-Based Meditation
In this study, “AI-based meditation” is defined not as an umbrella term for all forms of digital meditation technology, but more narrowly as interventional systems in which algorithms analyze aspects of the practitioner’s state and may shape the flow of experience. While such systems appear in diverse forms—including content-delivery applications, biosignal-based biofeedback platforms, conversational AI meditation coaches, and immersive virtual-reality environments (see
Appendix A for an overview of representative AI-mediated meditation systems considered in this study), the present analysis is restricted to those that share three defining features.
Table A1 organizes these systems into five types according to their representative function, mode of intervention, and concrete operational examples. For the purposes of the theoretical analysis developed in this study, the five types are not treated as distinct cases requiring separate analysis. Rather, they are understood as instantiations of a common structural logic: each datafies some aspect of the practitioner’s state, delivers feedback through algorithmic processing, and organizes practice into procedurally structured sequences. The variation across types is not irrelevant—it indicates that pre-configured self-understanding is not a single mechanism but a structural tendency that manifests at different levels of technological mediation and with varying degrees of conditional concealment. In what follows, the analysis draws on this typology to illustrate how the three defining features identified above operate across different system architectures.
A biofeedback platform illustrates all three: it datafies physiological signals into emotional categories, delivers feedback through algorithmic classification, and organizes practice into procedurally sequenced stages determined by system outputs rather than practitioner observation. These features are: (1) the datafication and classification of the practitioner’s state; (2) the delivery of feedback through algorithmic processes; and (3) the organization of practice into procedurally structured sequences. Together, these characteristics indicate a shift in the role of AI—from an auxiliary tool that supports practice to a mediating system that intervenes in the arrangement of the conditions through which experience is constituted. Accordingly, this study does not seek to generalize across all forms of AI-based meditation. Rather, it focuses specifically on those systems in which the conditions of experience are technologically pre-arranged. By delimiting the scope in this way, the analysis aims to examine how such configurations of conditional arrangements reshape the formation of experience and self-understanding.
What distinguishes AI-mediated meditation from earlier forms of digitally guided practice is not simply the presence of technological assistance or structured guidance, but the adaptive reconfiguration of experiential conditions in real time. Conventional digital meditation applications—such as audio-guided timers or pre-recorded instructional sequences—deliver fixed content that remains uniform regardless of the practitioner’s state. By contrast, the AI-mediated systems examined in this study dynamically reorganize the conditions of sensation, attention, and interpretation in response to continuously analyzed practitioner inputs. It is precisely this capacity for real-time, state-responsive pre-arrangement of experiential conditions that constitutes the specifically AI-driven dimension of “pre-configured self-understanding”—distinguishing it from digitally mediated contemplative environments more generally, in which the configuration of conditions, though technologically delivered, remains static and practitioner-independent.
3.2.2. The Conditional Structure of Self-Understanding in Traditional Meditation
Within the dependent origination framework, self-understanding is not an independent capacity residing within the individual. It is a process that forms through the interdependent combination of multilayered conditions—sensation, attention, emotion, and environment. Within this framework, the body is understood not as a fixed biological substrate or stable foundation of the self but as one of the interdependent conditions through which contemplative experience and self-understanding continuously arise, transform, and cease. Bodily sensation therefore functions as an important experiential site through which practitioners observe the dynamic interplay of conditions within meditative practice.
This understanding of the body as a condition through which experience becomes observable finds resonance in both phenomenological and enactivist perspectives. For
Merleau-Ponty (
1962, pp. 84–102), the body is not an object within experience but the very medium through which the world first becomes perceptible; the practitioner does not first exist and then perceive, but is already bodily oriented toward a perceptual field before any reflective act. This account converges with the satipaṭṭhāna framework, in which bodily sensation serves as the primary gateway through which conditions are encountered and observed, rather than as a merely instrumental substrate for mental activity. Similarly, the enactivist understanding of experience as emerging through ongoing reciprocal interaction between organism and environment (
Varela et al. 1991) reinforces the view that contemplative experience is relationally constituted rather than generated by an isolated inner subject.
Buddhaghosa’s Visuddhimagga distinguishes two complementary modes of practice: samatha (止, tranquility) and vipassanā (觀, insight). Vipassanā is described as the direct observation of the arising and cessation of the five aggregates (pañcakkhandha)—form (rūpa), feeling (vedanā), perception (saññā), mental formations (saṅkhāra), and consciousness (viññāṇa) (
Buddhaghosa 2010). This structure enables practitioners to observe in real time the conditions that constitute their experience and to witness directly how those conditions combine, transform, and dissolve. Traditional meditation, understood this way, can function as a practice environment in which conditions become more available for observation. Although contemplative practice is also shaped by teachers, doctrines, and institutional forms, practitioners are generally encouraged to engage with changing experiential conditions as they emerge within practice.
What this structure makes visible, repeatedly and concretely, is the non-linear way in which subtle changes in conditions produce different experiential outcomes from the same stimulus. Tranquility (pīti, sukha) arising during practice is not a static outcome of meditative proficiency or fixed personal disposition. It is a conditional phenomenon—temporarily arising when bodily tension, environmental stability, the locus of attention, and emotional background converge in a particular way. Conversely, the dispersal of concentration or the onset of bodily discomfort is not a failure in any simple sense; it reflects the way subtle shifts in conditions redirect the entire experiential field.
Traditional meditation reveals experience as a process constituted through changing configurations of conditions rather than a product of a single cause, allowing practitioners to observe how experience emerges through the interaction of sensation, attention, emotion, and context. It is important to acknowledge, however, that traditional meditation is not itself a structurally neutral or condition-free environment. The vipassanā tradition, for instance, operates within a normatively organized framework: the teacher’s authority structures the practitioner’s interpretive horizon, the sequence of practice stages prescribed in the Visuddhimagga imposes a specific directionality on experiential exploration, and the ethical regulations of the Vinaya establish prior conditions that shape the practitioner’s relational and behavioral field. In this sense, traditional meditation is not simply an open or unconditioned structure. What distinguishes it from AI-mediated meditation is not the absence of pre-arranged conditions but rather the degree to which the organization of those conditions remains relatively available to practitioners’ observation. In traditional contemplative environments, even the normative structures of teacher-guidance and staged practice are typically made available for reflective examination within the practice itself—the practitioner is invited to observe how these conditions operate, rather than to accept their outputs unreflectively. The analytical contrast developed in this study should therefore be understood not as an opposition between an idealized open form and a deficient closed one, but as a structural differentiation along a continuum of conditional visibility: from environments in which the configuration of conditions is relatively more accessible to the practitioner’s observation to those in which it is relatively more concealed within technological mediation.
3.2.3. The Dissolution of the Essentialist View of Self and the Emergence of Contemplative Agency
The rendering visible of conditions in traditional meditation leads practitioners away from the essentialist assumption that emotions and thoughts originate in a fixed inner entity. Instead, practitioners come to redefine these as cognitive constructs that manifest when multilayered conditions—sensation, attention, emotion, and environment—interact at a given moment. The fundamental orientation of traditional meditation shifts: rather than seeking a stable answer to the question “who am I?”, practitioners develop the capacity to observe the conditional structure through which the experience of “I” is generated. This resonates with the contemplative ideal of 隨處作主 (suicheo jakju)—meeting each situation as master of oneself, whatever the conditions one encounters.
The soteriological significance of this reorientation is clear. Traditional meditation enables practitioners to experience directly that their experience arises within a specific combination of conditions. Through this, it cultivates the capacity to recognize and reconstitute the conditional structure through which the self forms—what this study calls contemplative agency. This expands well beyond the traditional notion of a fixed essential self, offering the theoretical foundation for reconceptualizing contemplative agency as the capacity to critically observe and adjust the conditions that constitute one’s experience.
3.2.4. Conditions of Self-Understanding in AI-Based Meditation: Technological Intervention and the Normativity of Condition Configuration
AI-mediated meditation differs from traditional meditation not because it introduces mediation for the first time but because it may relocate some aspects of condition-formation into algorithmic and interface-based processes. It forms a distinct conditional structure because the conditions constituting experience are pre-set by technological algorithms from the outset. Before practitioners can explore their own sensation or emotion, they receive what is framed as objective data: a technological diagnosis derived from biometric signals and multimodal analysis—voice, facial expression, and related inputs (
Khan et al. 2025). The starting point of experience is thus shifted from the practitioner’s phenomenological self-observation to the normative labeling provided by the technology. Within this structure, emotional and attentional states may become increasingly organized through the classificatory logic of the algorithmic system rather than emerging solely through autonomous interpretation.
Having traced the conditional structure of AI-mediated meditation in Buddhist philosophical terms, it is now useful to approach the same structure from within philosophy of technology—not to replace the Buddhist framework, but to bring into view a dimension that it does not itself foreground: the tendency of technological systems not merely to arrange conditions but to conceal the fact of that arrangement. This structure bears formal resemblance to Gestell,
Heidegger’s (
1977) account of the essence of technology. Gestell is not simply the use of entities as tools; it is a prior determination of how entities will show up and be understood. In this respect, Gestell can be compared with the conditional structure of dependent origination, insofar as both involve the pre-arrangement of the conditions under which beings appear. But an important divergence separates them. Dependent origination aims at dissolving grasping and reification by making interdependent arising transparent. Gestell, by contrast, tends to conceal the operation of condition configuration and to reduce beings to a single interpretive possibility—what Heidegger calls “standing-reserve” (Bestand). The “pre-configured self-understanding” this study proposes refers precisely to this concealed precedence: in AI-based meditation, the algorithm partially organizes in advance how practitioners perceive their own experience by datafying and classifying their emotions and states before practice unfolds.
A potential concern within such environments is that practitioners may become more likely to rely on algorithmically mediated interpretations when outputs are presented as objective, data-driven, or diagnostically authoritative. The concern here is not that practitioners passively accept such interpretations, but that technologically mediated environments may influence how experiential states are categorized and made available for reflection. Some evidence suggests this is not merely theoretical. Studies have reported that algorithmic outputs in technologically mediated contexts can exert significant influence on how users understand their own states. This study treats this not as a universal empirical finding to be generalized but as a theoretical tendency inherent in the structural conditions through which self-understanding forms in technologically mediated environments.
The technology’s language intervenes directly in how practitioners represent and assign meaning to their experience. It may guide practitioners to align themselves with the categories the system has established, rather than discovering meaning through their own observation. These tendencies suggest the possibility that interpretive processes within contemplative practice may become more technologically mediated, as practitioners increasingly engage with classifications, feedback systems, and externally structured interpretive frameworks. This possibility aligns with emerging findings from human–AI feedback research suggesting that technologically mediated outputs can influence how individuals engage with and interpret self-relevant information (
Glickman and Sharot 2025). These tendencies suggest that practitioners’ engagement with experience may increasingly operate within technologically mediated interpretive frameworks, as classifications, feedback systems, and interface design shape how experiential states are categorized, presented, and made available for reflection. The concern is therefore not that practitioner agency is eliminated but that interpretive processes may become partially redistributed within technologically mediated environments.
3.2.5. The Standardization of Practice and the Reduction of Self-Understanding in Fragmented Environments
AI-mediated meditation may also shape the temporal flow of contemplative practice through algorithmically structured procedures, such as voice interfaces, sequenced prompts, and feedback systems. These features can guide attention in ways that differ from more open-ended observation of arising sensations and emotions. Rather than eliminating practitioner agency, such structures may influence how practitioners move through contemplative processes and how experiential states become available for interpretation.
This dynamic may become more pronounced within fragmented everyday contexts characterized by interruptions, mobility, and competing attentional demands. In such environments, practitioners may engage more frequently with immediate feedback and predefined categories when interpreting their experiences. The concern is not that technologically mediated feedback necessarily prevents reflection but that it may shape the rhythm, pacing, and direction of contemplative engagement in structured ways.
Taken together, traditional and AI-mediated meditation can be understood as different configurations of mediation rather than oppositions between open and closed systems. Traditional contemplative settings often provide greater opportunities for observing conditions as they arise within practice, whereas AI-mediated environments may organize aspects of contemplative experience through classification, sequencing, and feedback mechanisms before practitioners engage with them reflectively. These differences suggest that self-understanding in AI-mediated environments may increasingly operate within technologically mediated interpretive frameworks rather than emerging solely through open-ended observation.
3.3. The Distinctiveness of Dependent-Originationist Analysis: Contrast with Actor-Network Theory
From a dependent-originationist standpoint, self-understanding is not a fixed entity. It is a phenomenal event manifesting within the interdependent combination of conditions—sensation, attention, emotion, physical environment, and technological mediation (
Sangiacomo 2025). As the combination of conditions changes, experience transforms or ceases. This is precisely why traditional and AI-based meditation tend to produce different modes of self-awareness: the former generally supports contemplative environments in which conditions remain relatively more accessible to practitioners’ observation and reflective engagement, whereas the latter may tend toward more normatively organized structures in which algorithmic labeling and directive interface design partially pre-configure the conditions under which experience is interpreted.
Actor-Network Theory (ANT) is often brought to bear on the complex mechanisms through which humans and technology combine to constitute contemplative environments, and for good reason. ANT effectively traces how agency is distributed and stabilized within networks of human and non-human actors (
Ryan et al. 2024). Its limitation, however, is that its primary focus on structural configuration and role distribution within networks leaves it poorly positioned to explain how individual experience specifically arises through the combination of conditions, or how it transforms and ceases as conditions subtly shift.
Dependent origination centers precisely on what ANT marginalizes: the arising, transformation, and cessation of experience as functions of compositional changes in conditions. This allows the real-time tracking of how technological intervention reconstitutes a practitioner’s cognitive and emotional conditions and transforms self-awareness—something that ANT’s network-stabilization focus cannot easily accomplish. It is worth acknowledging that expanded approaches within ANT do attend to micro-level experience and practice. Still, the arising–transformation–cessation framework of dependent origination is better suited to the specific analytical purpose of this study. In this respect, dependent origination provides a distinctive framework for examining how the conditions through which self-understanding becomes intelligible are themselves continuously constituted and transformed within AI-mediated contemplative environments.