Three phenomena, three locations, one cause.
"When the chi passes through the occipital region of the brain people tend to see visions, and when it passes through the chi channels in the ear you tend to hear sounds including voices."1
Plus a third, from the same passage: chi ascending "sometimes producing clicking sounds, such as in the back of the head."1
Occipital region → visions. Ear channels → sounds and voices. Back of the head, ascending → clicks.
Which is a straightforwardly mechanical account. Not messages, not attainments, not contact with anything. The experience is determined by which region the chi is currently moving through, the way pressing your eyelid produces light.
And the occipital assignment is anatomically apt: the occipital lobe is where visual processing happens, so chi activity in that region producing visual phenomena is a claim that lines up with where vision is actually generated. The ear-to-sound mapping is the same logic.
The instruction is one line and it admits no qualification.
"You should always ignore such phenomena because they are illusions."1
Always. Not sometimes, not after evaluation, not once you have determined which are genuine.
The Zen position is invoked as the correct one: these things are illusions, or mara, and practitioners are told to disregard them — with the note that "people lacking deep cultivation knowledge" do not realise this.2
Which fits the mechanical account exactly. If a vision is what chi in the occipital region feels like, there is nothing in it to interpret — it is a byproduct, not a signal, and the same is true of the voices and the clicking.
The standing position on all such content is the same. See The Visions Are Fictitious.
The content, as opposed to the occurrence, has a stated source.
Nan Huai-chin's Tao and Longevity is cited as explaining this clearly, "including the fact that they arise from deep psychological tendencies."3
Which pairs with the wider account: the material is assembled out of "subconscious memories or knowledge of cultural symbolism."4
So there are two separate questions, and keeping them apart is the whole practical value here:
Why is anything happening? Chi in a particular region.
Why is it this? Your own psychology and cultural stock.
Neither answer gives you a reason to attend to the content. The first says the event is plumbing; the second says the imagery is autobiography.
🚩 SECONDARY WITHOUT PRIMARY — the Nan Huai-chin citation gives no page.
The reason the instruction is always ignore rather than discern carefully is a stated limit on what a practitioner can do.
"Only a master will know for sure, but regular human beings will not be able to differentiate between a projection in consciousness and genuine sensory input from the five senses."6
Projections and real perception are phenomenologically identical to you. There is no marker, no telltale, no quality of realness that distinguishes them.
Which makes any policy requiring discrimination unexecutable. A rule you cannot apply is not a rule, and the only workable option left is a blanket one.
And the blanket option costs nothing on this account, because nothing of value is being discarded. If the occurrence carries the only information — that chi is moving through a region — and the content carries none, then ignoring the content loses nothing.
Which is why the same conclusion is reached from two directions: the mechanical account says the content is a byproduct, and the epistemic limit says you couldn't sort it anyway.
When something clicks in the back of your head, note it and carry on. It is described as "just something you need to know sometimes happens" — nothing more is required of you.
When visions arrive, do not look closer. The instruction is to ignore, and looking closer is the opposite of ignoring.
When voices arrive, do not listen for content. Same category, same instruction, and it is reinforced at the more difficult stages where voices are explicitly never to be believed. See The Testing Maps the Blockages.
What you can usefully record is location and timing. Back of the head, ear, behind the eyes; during sitting, on waking, at a particular stage. That is the part with information in it, because on this account it maps to where the chi is working.
And keep practising through it. None of these is a reason to stop, adjust, or investigate.
One plain note: voices and visions outside of practice, or ones you cannot set aside, are also how serious illness presents. Nothing here distinguishes them, and that is a matter for a doctor first.
Worth setting against the phenomena, because the contrast is the point.
The achievement described at this stage is unspectacular: some degree of mental quiet, reachable "just as you can also achieve yourself, by following the simple path of meditation" — needing only "to put time into just watching thoughts go by."10
And the verdict attached: "this accomplishment, rather than his feats of strength, is what makes Hercules a hero worthy of our admiration."10
It is also explicitly modest in scale — a mundane stage rather than a samadhi attainment, "in Buddhism a minor calming of the sixth consciousness."11
Which is the useful frame for everything above. The clicks and lights and voices are loud; the actual accomplishment is quiet and small — and a practitioner grading themselves on phenomena is measuring the byproduct while the thing being built makes no noise at all. See Gong Fu Is the Signpost, Not the Road.
The material points repeatedly at one book for this territory.
Tao and Longevity by Nan Huai-chin — cited for the psychological origin of these phenomena,3 for the jing-to-chi scheme,7 and as the standing reference: "the world is extremely lucky to have Tao and Longevity, by Nan Huai-chin, that explains many of the stages of the path using Buddhist, Confucian and Taoist terminology."8
With the division of labour stated: that book covers the stages; the present material covers the Vajrayāna generation and completion stages, before and well after the sushumna opens.8
Warrant flag: the author is this practitioner's teacher, and the recommendation is not independent. The vault records the pointer and the relationship together.
There is a fourth category in the same family, and it is the one with no handling instruction attached.
Memories and urges surface, tied to specific chi routes, with corresponding reactions in consciousness.9
Which is the same mechanism — chi moving through a region produces an experience — applied to psychological rather than sensory content.
And nothing is said about what to do with it. The clicks, visions and voices get always ignore. Surfacing memories get described and left there.
Which matters because they are the category most likely to demand a response. A click does not ask anything of you; a surfacing memory does. See The Testing Maps the Blockages.
Evidence. [PRACTITIONER SOURCE]. 🚩 SINGLE SOURCE 🚩 SECONDARY WITHOUT PRIMARY. [LOW CONFIDENCE] on the mechanism — the location-to-phenomenon mapping is asserted, not demonstrated.
Tension — ignore, and track. These phenomena are to be always ignored, and their location is the one thing that would tell you where the chi is working. A marker you must disregard is not usable as a marker.
Tension — illusions, and beings. Here the account is purely mechanical; elsewhere the same class of content is attributed to projections deliberately supplied by other beings.4 Both are given, and they are different claims about what a vision is.
Open question — what about surfacing memories? The one category that plausibly needs a response is the one with no instruction.
Open question — does the mapping extend? Occipital, ear, and the back of the head are named. Whether other regions produce their own characteristic phenomena is never addressed, and it is the obvious next question.
Two accounts of the same vision, held in the same material.
A byproduct. Chi passes through the occipital region and you see something, the way pressure on the eye produces light. Nobody sent it, it means nothing, and the content comes from your own memory.
And a message that isn't one. Projections supplied by other beings, deliberately, out of compassion, drawing on your cultural stock to keep you moving.
The practical instruction is identical under both — ignore it — which is presumably why the difference is never resolved.
But they are not the same claim, and they differ on something a practitioner might care about: whether anything is paying attention to them. The mechanical account says no. The projection account says yes, and that the display is authored.
Hebb's Sensory Deprivation Experiments — psychology.
When ordinary sensory input is removed, subjects reliably begin to hallucinate — first simple geometric forms and lights, then complex scenes, then sometimes voices. The findings are old, robust, and produced in healthy volunteers with no expectation of anything unusual.
Two things carry over.
Reduced input produces manufactured content. The system does not idle when it is starved; it generates. And the progression is orderly — simple phenomena first, elaborate ones later, which suggests the content is being built up rather than received.
And the phenomena are not pathological. Ordinary people in a quiet room for long enough produce them. That is what the machinery does under those conditions.
Which describes a meditator's situation with uncomfortable precision. Eyes closed, still, quiet, for hours is a controlled reduction of sensory input — and the phenomena reported at these stages are exactly the deprivation sequence: clicks and simple sounds, lights and geometric forms, then figures and voices.
That makes the ignore them instruction better founded than the material's own reasoning supports. You do not need a channel theory to expect these. Anyone sitting still with their eyes shut for long enough is running the experiment, and the content that arrives is what the system produces under input starvation, assembled — as both accounts agree — from the person's own stock.
It also explains the reassurance about practitioners who get nothing. Individual variation in the deprivation literature is enormous, and it tracks the person rather than the depth of anything.
Practical form: when phenomena appear under conditions that reliably manufacture phenomena, their arrival is information about the conditions and not about you — which is the strongest available reason to do exactly what the tradition already says: note it, and carry on.
What the source leaves undetermined is whether the location mapping is worth anything.
The claim is precise and, if true, useful: visions mean chi in the occipital region, sounds mean the ear channels, clicks mean the back of the head. That is a position readout — a way of knowing where the work currently is without seeing anything.
And the instruction is to ignore all of it.
So the one genuinely diagnostic feature of these phenomena — their location — is dismissed along with the content that isn't diagnostic. Nothing distinguishes the two in the instruction.
Which leaves an obvious refinement unstated: disregard what you see, record where it came from. That would keep the safety of the blanket rule while preserving the only information the phenomena carry.
The material has both halves — the mechanism and the prohibition — and never puts them together. Which is the difference between a practitioner who knows roughly where their chi has reached and one who has been told to look away from the only instrument that would tell them.