Three modalities travel to a remote client: something they watch, something they hear, and something they feel. Any one of them is a complete set. You choose based on the client in front of you, not on what the tool defaults to.

Remote adds two problems an in-office kit never had. The first is sync — if the tone arrives a beat behind the moving stimulus, the client is no longer receiving alternating bilateral input, and you cannot hear their audio from your side of the screen, so nothing will tell you. The second is stereo separation — audio bilateral stimulation only works if left and right are genuinely distinct, which means headphones, not the speaker on a phone.

Rate is the third thing worth setting deliberately. The standard protocol sits around one movement per second. Faster multimodal variants published in the research literature work in a range up to roughly two. Nothing published goes past two, and above about three flashes per second you are into territory the web accessibility guidelines treat as a seizure-risk surface. Two is both the top of clinical practice and comfortably below that line.

Which modality for which client?

Visual is the default and the most familiar — a shape moving across a field. It is also the only modality carrying photosensitive-seizure considerations, so it is the one to switch off rather than negotiate when a client screens positive for photosensitive epilepsy, recent concussion, migraine with aura, a vestibular disorder or recent eye surgery.

Auditory is a tone that moves between the ears. It carries no photosensitive risk and works for clients who find the visual overwhelming or who close their eyes to reprocess. It fails silently on mono output, which is the single most common preventable problem in a remote EMDR session.

Tactile is a tap that alternates between left and right. It is the one that most closely resembles the in-room buzzers, and the one with the most variation in how it gets delivered remotely.

A client who is overwhelmed by two modalities at once is telling you something. Drop to one. Multimodal stimulation is a feature of certain protocols, not a requirement of the therapy.

What rate should I use?

Set it by protocol and then by the client.

• Around one movement per second is the standard-protocol baseline and the rate most practitioners were trained on.

• Slower — well under one is useful for early tolerance testing, for resource installation, and for clients who report the movement itself as aversive.

• Faster, up to about two is where the multimodal variants sit. This is a protocol choice that follows training, not a dial to turn because a session feels slow.

If your tool lets you set a rate above two, that is worth noticing. Published EMDR protocols do not go there, and sustained flicker faster than about three per second is the threshold the accessibility literature treats as a photosensitive-seizure risk.

Adjusting rate mid-set is legitimate and sometimes clinically indicated. What you want is for the change to land without the sound sliding out of step with the picture.

Why does sync matter so much?

Because the whole mechanism is alternation, and because you cannot verify it from your side.

The client hears the tone through their own headphones. You do not. If the audio drifts a quarter-second behind the visual over the course of a thirty-second set — which is exactly what happens when two separate clocks drive two separate channels — the client is getting a smeared, roughly alternating stimulus and neither of you knows.

The practical test is cheap. In the first session, run a short neutral set, then ask directly: does the sound land where the shape turns, or slightly after? A client who has to think about it is telling you it is close enough. A client who says "the beep is behind it" is telling you to run one modality only until you have a tool that holds them together.

What about headphones?

Say it explicitly in the pre-session note, because it is the failure that looks like everything working.

In rough order of preference: over-ear or on-ear headphones, wired earbuds, Bluetooth earbuds, laptop speakers, and — last, and effectively useless for audio bilateral stimulation — a phone speaker. Most phones output mono from a single speaker, so the tone simply does not move.

Bluetooth earbuds work but occasionally hiccup during a reconnect, which shows up as a stutter mid-set. Wired is duller and better.

How long should a set run?

Practitioners vary, and the honest answer is that set length is a clinical decision you already make in the room. What matters more remotely is what happens at the end of one.

A set that ends and then waits for you is the right shape. A tool that starts the next set on its own has made a clinical decision it isn't in a position to make — the check-in between sets is where you find out whether to continue, slow down, change modality or move to closure.

What do I do if something goes wrong mid-set?

Stop the set. Not at the end of the timer — now.

That applies to a client reporting visual discomfort, headache, dizziness or "this feels wrong," and it applies just as much to a frozen picture. A stimulus that keeps running while you cannot observe your client is not a set you are running.

Then check in, reconfigure before the next set rather than during this one, and document what happened. If the trigger was visual, the next set runs on sound or touch. Neither carries photosensitive risk.

The short version

Pick modalities for the client, keep the rate inside the published range, insist on headphones if you're using audio, and verify once — out loud, in the first session — that the sound lands where the picture turns. Then watch their face, and stop the set yourself the moment you can't.

About Rivet

Rivet is a Canadian practice line built for therapists — a separate number for calls, texts and voicemail, with video sessions and clinical tools in the same place. Your data stays in Canada, and transcription runs on Rivet's own hardware rather than a third-party AI service.

One plan, $65 CAD a month, everything included. Fourteen-day trial, no card.

See what's included · Start the trial