Pratyāhāra: The Vedic Protocol Modern Attention Training Has Been Reinventing for Thirty Years

Tuesday, 19 May 2026

A fourteen-year-old came to my consulting room three days before her board examination. Her first sentence was specific. "I can hear everything in the room. I cannot stop hearing."

She described, in detail, what her senses were doing without her consent. The pen-tap of the student two desks away. The clock. The proctor's footsteps. The breath of the girl behind her. The smell of the room before an examination begins. The visual scan of who had reached page two while she was still on page one. Each sense was feeding her citta a stream of data that the meaning-making apparatus was then converting into evidence of her own inadequacy. By the time she sat down at her desk, the catastrophising loop had already been running for forty minutes.

This is the clinical engine that exam-anxiety literature in the last decade has been trying to name. It is not, primarily, interoceptive sensitivity. The cold hands and tight throat are downstream. The upstream event is the failure of the indriyas to stop coupling with their objects. The senses keep delivering. The citta keeps interpreting. There is no off-switch in the loop.

Modern psychology has built a protocol for this. The two best-developed strands are the Attention Training Technique (Wells, 1990, 2009), which sits at the centre of metacognitive therapy, and the decoupling or decentering construct that drives much of MBCT's relapse prevention effect (Teasdale et al., 2002; Bernstein et al., 2015). Both protocols teach the patient that attention is not a passive recipient of sensory input. The patient can decide, with training, what the indriyas are allowed to deliver to the citta.

The Yoga Sūtras described the same operation two thousand years before the first ATT trial, and described it more completely. The Sanskrit name for it is pratyāhāra.

What pratyāhāra actually is

Patañjali defines it in a single sūtra. Sva-viṣaya-asamprayoge cittasya svarūpānukāra iva indriyāṇāṃ pratyāhāraḥ (Yoga Sūtras II.54). When the senses are no longer coupled with their own objects, and they begin to imitate the inward orientation of the citta, that uncoupling is pratyāhāra. The next sūtra (II.55) names the result. Tataḥ paramā vaśyatā indriyāṇām. From this comes supreme mastery over the senses.

Three precise points are worth holding.

First, pratyāhāra is a withdrawal. The English translation "sense-control" softens what the Sanskrit is doing. The indriyas are pulled back from their viṣayas. The eye stops delivering visual data to the meaning-making apparatus. The ear stops delivering auditory data. Touch, taste, smell, the same. The senses are still functional. They are no longer feeding the engine.

Second, the withdrawal is voluntary and structural, not suppressive. The patient is not pushing the input away. The patient is learning that the indriya is permitted to be off-line. The clinical translation is that attention is the gate, and the gate can be closed.

Third, the senses, once uncoupled, follow the citta. They do not float free. They are now available to be deployed by the practitioner toward an object of the practitioner's choosing. This is why pratyāhāra is the fifth limb of Aṣṭāṅga and dhāraṇā is the sixth. Pratyāhāra makes single-pointed attention possible by clearing the channel.

This is the operation that modern attention training has been climbing toward.

What ATT and decoupling have got right, and what they leave incomplete

The Attention Training Technique is rigorous. Patient sits with eyes closed. The therapist or recording plays five to seven competing auditory stimuli at varying spatial locations. The patient is asked to selectively attend to one stream, then switch to another, then attend to all simultaneously without selecting. The mechanism is the strengthening of voluntary executive control over attention deployment, with the corollary that the patient experiences, often for the first time, that attention is a faculty they can operate.

The data is good. Knowles et al. (2016) found medium-to-large effect sizes for ATT across anxiety and depression samples in their systematic review. The technique is brief, structured, and replicable.

What ATT does not provide is what Patañjali insisted on. Substrate.

ATT begins with attention training. Patañjali's protocol begins four limbs earlier. Yama and niyama establish the ethical and lifestyle preconditions that reduce baseline arousal. Āsana stabilises the body. Prāṇāyāma regulates the breath and lowers sympathetic tone (Bernardi et al., 2001). Only on top of this floor does pratyāhāra become trainable. ATT often works without these substrate layers, and when it fails, it tends to fail in patients whose physiological baseline is too elevated for attentional control to be acquired. The substrate is not optional. It is what makes the technique trainable in the first place.

ATT also leaves implicit what pratyāhāra makes explicit. Pratyāhāra is not the goal. It is the gate through which dhāraṇā becomes possible. The patient who learns pratyāhāra and stops there has acquired a skill that decays without an anchor to deploy it on. Patañjali's sequence is pratyāhāra into dhāraṇā into dhyāna. The withdrawal is for the sake of the anchoring. ATT has no equivalent next-step, and clinicians who use it sometimes report patients who have the technique and find nothing to do with it.

This is the additive layer the Yoga Sūtras supply. Pratyāhāra is not a new name for ATT. It is the older, more complete protocol, and the difference is operational.

Cognitive behaviour therapy is not weakened by this observation

The point is not that CBT is inferior. Metacognitive therapy and MBCT are extensions of CBT and they are effective. Pratyāhāra clarifies the architecture inside which these interventions can be sequenced. CBT remains the load-bearing tool for cognitive restructuring. MCT remains the load-bearing tool for attention deployment. Pratyāhāra supplies the prior step that both CBT and MCT often skip. It also supplies the next step that ATT does not articulate. The framework is additive on both ends.

The clinical protocol

A working pratyāhāra sequence for an adolescent in the situation I have described looks roughly like this.

First, regulated breath. Three to five minutes of nāḍī śodhana or extended exhalation. Not relaxation. Arousal-reduction. Bernardi et al. (2001) measured the parasympathetic shift and it is reliable.

Second, attentional flexibility training. An ATT-style exercise adapted to her age. Three sounds in the room. Two minutes selectively attending to one. Two minutes switching. Two minutes dividing. This trains the executive faculty before pratyāhāra is asked of it.

Third, the pratyāhāra move. The instruction now changes. The patient is asked to withdraw all three streams. Not to attend to one. Not to switch. To allow the indriyas to be off-line, briefly, while remaining alert. The first time most adolescents do this, the experience is unmistakable. The room becomes quieter without the room becoming quieter.

Fourth, dhāraṇā. With the indriyas no longer feeding the engine, the attention is offered a single stable object. A mantra. The breath. The space between the eyebrows. The patient is now in the territory where the catastrophising loop can be interrupted at source, because the source is no longer being fed.

By the third session of this protocol, she could sit at her desk for forty minutes without the loop engaging. By the time she sat the examination, the room was still loud. Her citta was not being asked to interpret all of it.

What changed for her, in her words

She said one sentence at her last session that summarises the mechanism better than any review paper. "I can still hear the room. I do not have to listen to it any more."

That is pratyāhāra. The indriyas remain functional. The colonisation of the citta stops.

A note on diagnosis

Her clinical picture meets criteria for adolescent generalised anxiety with a strong performance-anxiety component. The DSM-5-TR formulation is accurate and useful. The pratyāhāra formulation is additive. It tells the clinician what specifically she needs to learn to do, in a way the diagnosis alone does not. The plan written in DSM language reads "attention training plus cognitive restructuring." The plan written in Patañjali's language reads "build substrate, train flexibility, withdraw the indriyas, anchor on a single object." Both are correct. The second tells the therapist exactly what to do on Monday at four o'clock.

Closing

The contemporary anxiety literature has been climbing toward Patañjali's account from below for thirty years. ATT and decoupling are the closest contemporary protocols. Pratyāhāra is the older, more complete version, and it supplies both the substrate the modern protocols often skip and the next step they do not articulate.

If this resonates, and you would like to explore DVT's potential in improving your mental wellbeing, book a consult: anandohamhealth.com/book

References

Bernardi, L., Sleight, P., Bandinelli, G., Cencetti, S., Fattorini, L., Wdowczyc-Szulc, J., & Lagi, A. (2001). Effect of rosary prayer and yoga mantras on autonomic cardiovascular rhythms. BMJ, 323(7327), 1446–1449.

Bernstein, A., Hadash, Y., Lichtash, Y., Tanay, G., Shepherd, K., & Fresco, D. M. (2015). Decentering and related constructs: A critical review and metacognitive processes model. Perspectives on Psychological Science, 10(5), 599–617.

Knowles, M. M., Foden, P., El-Deredy, W., & Wells, A. (2016). A systematic review of efficacy of the Attention Training Technique in clinical and nonclinical samples. Journal of Clinical Psychology, 72(10), 999–1025.

Patañjali. Yoga Sūtras, II.54, II.55. (Multiple translations consulted.)

Teasdale, J. D., Moore, R. G., Hayhurst, H., Pope, M., Williams, S., & Segal, Z. V. (2002). Metacognitive awareness and prevention of relapse in depression. Journal of Consulting and Clinical Psychology, 70(2), 275–287.

Wells, A. (1990). Panic disorder in association with relaxation-induced anxiety: An attentional training approach to treatment. Behavior Therapy, 21(3), 273–280.

Wells, A. (2009). Metacognitive therapy for anxiety and depression. Guilford Press.