Thermal feedback is slow to respond, unsuitable for instant confirmation

What it is

From the moment a thermal actuator starts heating or cooling to the moment a user actually registers "the temperature changed" and can act on it, far more time passes than with vibration or pressure feedback — typically hundreds of milliseconds to several seconds. That makes temperature an inherently slow feedback channel, unsuited to anything requiring instant confirmation, such as acknowledging the moment a button is pressed; even making the temperature change larger cannot fundamentally fix this latency.

Why it happens

The delay comes from two layers stacking. The first is physical: a thermal actuator has thermal inertia, and heat takes time to conduct through the contact surface down to where the relevant receptors sit in the skin, so the actual skin-temperature curve lags noticeably behind the drive signal — a lag on the order of hundreds of milliseconds to seconds on its own. The second is neural: temperature signals are carried by relatively thin, slower-conducting nerve fibers, much slower than the thicker, faster fibers carrying vibration or pressure signals — so even once skin temperature has genuinely changed, the time for that change to be encoded by receptors, transmitted centrally, and registered consciously as "felt" is longer than for a vibration signal over the same timeline. The two delays compounding is why the thermal channel's total time from "physical change begins" to "user can react" runs noticeably longer than other tactile channels.

Studying it

A common approach has participants perform simple detection or reaction-time tasks separately for thermal, vibrotactile, and pressure cues, while sensors such as contact thermocouples measure the physical response curve between the actuator's drive signal and the actual skin-contact temperature — separating "physical response delay" from "behavioral response delay" so the source of the overall lag can be identified.

Where it stops holding

The exact size of the delay varies with the magnitude of the temperature change and the quality of contact between actuator and skin — a bigger delta and poorer contact both make the lag more pronounced. Some fast-onset strong cold stimuli can sometimes be detected faster than slow warming, but even the fastest thermal cue remains noticeably slower to register than vibrotactile feedback. This doesn't mean thermal feedback is useless — only that it isn't suited to a confirmation task with a hard real-time requirement (on the order of a hundred milliseconds).

Applying it

• Don't make thermal feedback the primary channel for anything needing rapid confirmation (button-press acknowledgment, an error alert that must be noticed immediately) — use vibration or pressure feedback for that, and let a thermal cue arrive afterward as a secondary layer if needed.
• Reserve thermal feedback for states that already unfold slowly, on the order of seconds (ambient comfort adjustment, charging progress) — situations with no requirement for an instant reaction — so its response speed matches the pace of what it's conveying.
• How to check: for the actual hardware in use, measure end-to-end latency from drive signal to user-reported detection, and compare it against the maximum delay the task can tolerate, to confirm whether the thermal channel meets that specific task's real-time requirement.

Related

• Same group: A4.06.1 Thermal perception is more sensitive to the rate of change than to absolute temperature · A4.06.3 Device warmth gets misread as a malfunction signal
• Nearby: A4.15 Thermal perception and thermal feedback · A4.17 Temporal resolution and synchrony windows in touch
• Search terms: thermal feedback latency · thermal reaction time · C fiber conduction · haptic channel comparison