Context-dependent memory explains why switching devices carries a nonzero relearning cost

What it is

A user who has learned an app on their phone and then tries to perform "the same feature" on a tablet or desktop usually still needs a moment to readjust, even when the underlying logic is identical on both ends. This isn't carelessness or lack of skill — it's that the device itself is part of the memory context. Screen size, input method (touch versus mouse), and the concrete layout of the interface were all part of the context encoded alongside the action, and switching devices changes that part of the context, which is why the relearning cost is never actually zero.

Why it happens

A user's memory for an action binds together more than the abstract task logic (which control produces which outcome) — it also binds the specific sensorimotor details involved in performing it: where a finger swiped on the screen, the distance and direction a mouse traveled, how elements were arranged relative to each other at a particular screen ratio. Moving to another device leaves the logical goal unchanged, but the contextual cues that supported retrieval — input method, spatial layout, perceptual detail — no longer match, leaving fewer matching signals to draw on. Making up the difference takes a bit of extra attention, which is exactly the experience of "I've used this before, but I still have to feel my way through it on a new device."

Where it stops holding

The cross-device relearning cost isn't fixed: when the two devices share a highly consistent interaction model — same input method, similar spatial proportions, a consistent visual language — the contextual overlap is high and the cost approaches zero. When the input method itself changes qualitatively — touch gestures replaced by mouse clicks, or voice replaced by keyboard — the mismatch isn't just visual but sensorimotor, and the cost is noticeably higher. The cost also shrinks with experience: users who switch between multiple devices frequently gradually abstract their operating knowledge into a representation that doesn't depend on device-specific details, weakening the context-dependence over time.

Applying it

• When designing a cross-device experience, don't assume "consistent logic" means "zero learning cost." Measure time-to-completion the first time a user performs the same action on a new device as an independent metric of cross-device experience quality, rather than assuming it matches same-device performance.
• Keep core cues consistent across devices wherever possible — the same icon, a similar relative position, consistent terminology — to shrink the scope of the mismatch, and prioritize preserving a handful of highly recognizable anchors even when layout has to change for screen size.
• For cases where the input method changes qualitatively (touch to desktop mouse-and-keyboard, for instance), provide device-specific guidance rather than reusing onboarding content written for the other platform.
• Verification: find users already proficient with a feature on device A, time their first attempt at the same action on device B, and compare that against new users native to device B. Similar times mean the cross-device design has largely eliminated the context mismatch; a noticeably slower time means the design is still leaning on device-specific cues.

Related

• Same group: A6.08.1 Retrieval is easier when cues match the encoding context · A6.08.2 The interface itself can serve as an external retrieval cue · A6.08.3 Changing an interface removes a user's memory cues · A6.08.5 Internal states such as mood and arousal are also part of the context · A6.08.6 Context-dependence affects recognition tasks less than recall tasks
• Nearby: A6.09 Procedural memory and automatization
• Search terms: context-dependent memory · cross-device consistency · platform switching cost