A social robot that provides no lasting practical value gets abandoned once the novelty fades

What it is

Once the novelty effect fades, what determines whether a user keeps using a social robot is whether it keeps solving a real need — its actual utility. If a robot has nothing to offer beyond expressing emotion and being interactive, it is likely to be abandoned once the novelty wears off: left in a corner, unused except to show visitors.

Why it happens

This follows directly from the decline of the novelty effect. Once interaction no longer pays out the exploratory reward that novelty provided, users re-assess the opportunity cost of continuing to invest time, space, and attention — and that re-assessment is no longer based on "is this fun" but on "what problem does this actually solve." Pure affective expression or companionship value, unless it is paired with a genuine functional need (medication reminders, health monitoring, household help, tutoring support), tends to lose that re-assessment, because whether companionship keeps being perceived as valuable depends heavily on whether the user actually needs companionship at that moment — something a designer cannot dictate unilaterally. The robot ends up idle not because it broke, but because its value proposition does not survive the moment novelty is gone. It is worth distinguishing that affective companionship value itself decays over time and needs continually refreshed content to stay engaging — the same joke or the same emotional display loses its pull after enough repetitions — unlike a stable functional task such as a medication reminder, whose value does not depend on freshness, only on whether the task still exists. This is why simply stacking on more affective content rarely offsets abandonment risk for long: it is exhausted by the same mechanism, just more slowly than pure exploratory novelty.

Studying it

A common approach is to track, within a long-term deployment study, both the abandonment rate (what fraction of users stop using the robot, and within what timeframe) and when abandonment occurs, combined with interviews asking users for the specific reason — sorted into categories such as "not useful enough," "maintenance burden too high" (charging, cleaning, error recovery), or "no longer novel" — to test whether functional value predicts retention. A second approach compares retention curves between task-oriented deployments and purely companionship-oriented deployments, checking whether the two curves diverge once novelty has decayed. Reuse-triggering events can also serve as an independent variable — the rate at which an idle device gets reactivated after a new skill is pushed, firmware is updated, or a family member demonstrates it again — which directly distinguishes disuse from forgetting the device exists versus disuse from an active decision to stop using it, since each calls for a different intervention.

Where it stops holding

If a product is explicitly positioned from the outset as a one-off or short-term affective or entertainment good — a holiday-themed interactive toy, a robot rented for a children's birthday party — then being abandoned after a few uses is not a failure at all; this concept only holds for products that claim to offer long-term companionship or assistant services. Utility also has to be judged by value the user actually perceives, not by what a designer assumes is useful — a feature a designer considers highly practical still leads to abandonment if users do not feel they need it. Research-provided hardware, with free replacement and a team on call for repairs, also systematically understates commercial maintenance costs and purchase-price barriers — when users pay for and maintain the device themselves, the threshold for abandoning it is much lower than in a research deployment, so retention measured in a lab or grant-funded study should not be treated as a forecast for the retail setting.

Applying it

At the design stage, identify at least one real need the robot can keep meeting repeatedly without relying on novelty — a schedule reminder, health monitoring, household assistance — and keep exposing that function to the user throughout long-term deployment, rather than demonstrating it once during unboxing and never surfacing it again. The concrete check is to interview users around the point where abandonment signs appear and ask directly: "if we removed its emotional expression and companionship features, would you still use it?" If most answers are no, the product currently lacks the real utility needed to sustain long-term retention, and the gap needs to be closed at the functional level, not the affective one. An idle device also needs a low-cost path back to use — a one-tap reset that pushes the latest content pack, or a remote nudge a family member can trigger to demonstrate a new feature — so reactivation costs close to nothing instead of requiring the user to redo initial setup from scratch.

Related

• Same group: X6.04.1 Early enthusiasm for a social robot does not predict whether people keep using it · X6.04.3 Judging whether a social robot succeeds requires evaluating it over a long deployment, not a short trial
• Nearby: X6.05 Use by special populations · X4.07 Operator situation awareness
• Search terms: utility · robot abandonment · long-term HRI · novelty effect