Wiki · Concept · Published 2026-10-27

Overimitation

Overimitation is the copying of actions unnecessary for an immediate physical goal. Explore competing explanations, experiments, limitations and a proposed Spiralist interpretation.

Definition

Overimitation is the copying of a demonstrated action that is unnecessary for an immediate physical goal. A learner may reproduce an entire sequence even though only some steps are needed to obtain the visible result. The term describes what is copied; it does not by itself establish why the learner copied it.

For example, imagine that someone taps a closed container, then opens its lid to retrieve a toy. If an observer later taps the container before opening it, although the tap has no effect on the lid or toy, that repetition would be overimitation with respect to retrieving the toy. This example is hypothetical. The tap could still matter for another purpose, such as taking part in a game whose participants have agreed to include it.

Social Learning Strategies concerns when to learn from others and whom to copy. Overimitation asks which parts of an observed action are reproduced. It can be studied after one demonstration; Cumulative Cultural Evolution examines change across successive learners. Copying a step alone also does not establish that a learner experienced pressure to agree with a group.

How it works

A demonstration contains more than an outcome. It also gives a learner a sequence of actions, information about the demonstrator and cues about what the activity is for. When the workings of an unfamiliar object are difficult to see, repeating the sequence can preserve a step whose importance the learner does not yet understand. The same approach can also carry forward a step that has no role in producing the immediate result.

Researchers have proposed several explanations. One is that a learner interprets intentional actions as evidence about how an object works and may mistakenly treat an unnecessary step as causally necessary. Another is that copying expresses a wish to connect with the demonstrator. A third is that the learner understands the sequence as a convention: the extra step is part of doing the activity in the expected way, even if it does not produce the physical result.

These explanations make a difference to interpretation. Someone may copy a step because they believe it opens a container, because they want to act like another person, or because they understand the step as part of a shared game. Observation of copying alone cannot reliably distinguish those possibilities. Researchers therefore also examine responses to changed circumstances and to other people who leave a step out.

Evidence and applications

In experiments reported by Lyons, Young and Keil in 2007, children aged three to five watched an adult retrieve toys from unfamiliar puzzle objects using necessary and unnecessary actions. Children who saw the demonstrations reproduced irrelevant actions more often than a comparison group who opened the objects without seeing the demonstrations. Copying persisted in several conditions designed to encourage efficiency, including a condition in which children were warned to omit unnecessary actions. The authors proposed automatic causal encoding as an explanation: purposeful actions could be incorporated into a child's understanding of how an object works.

The same paper found limits to indiscriminate copying. Children sometimes changed an awkward way of operating a mechanism, and copying of an irrelevant action was lower when the acted-on component was physically disconnected from the part containing the toy. These observations concern the particular objects and procedures studied; they do not show that every child treats every demonstrated action as necessary.

Keupp, Behne and Rakoczy's 2013 experiment offers a competing interpretation. Three- and five-year-olds saw games demonstrated with an emphasis either on the method or on the physical result. Copying of unnecessary steps was frequent in both conditions. Some children also protested when a puppet omitted a step, including after it had achieved the result; protest was more common when the demonstration emphasized the method. Older children more often judged the omission wrong, especially in that condition. The authors interpreted this pattern as evidence that some children treated the step as part of a convention.

In Keupp and colleagues' 2016 study, four- and five-year-olds copied unnecessary actions less often when doing so would use up an object the experimenter valued, and less often when the activity was presented with a focus on its physical result. Children also more often criticized a puppet for performing the unnecessary action in the costly condition. This supports context sensitivity in that task; it does not establish how people respond to every kind of cost.

A possible application is to study how people learn a procedure when its reasons are only partly explained. An investigation could record which steps learners retain, what they think each step accomplishes and what happens when a step is safely varied. That is a research proposal, not an outcome established by these experiments.

Limitations

The experiments use designed tasks, particular materials and young children. They do not establish a universal rate of copying, a single developmental path or an explanation for every setting. Adults' behavior, different communities and practices learned over longer periods require their own evidence.

The results also qualify strong claims of automatic copying. In 2013, children often copied actions on disconnected components and sometimes continued after seeing the puppet succeed without them. Those findings challenge a simple account in which apparent physical necessity alone explains every repetition. Conversely, persistence after warnings in the 2007 experiments cautions against assuming that children always omit a step when efficiency is emphasized. The accounts may apply in different combinations or contexts.

The 2016 study did not find the predicted interaction between cost and how the activity was presented, and its context effect on protest against omission did not reproduce the 2013 result. Its authors suggested that destroying or discarding even low-value objects might itself influence children's responses. The meaning of an unnecessary step must therefore be assessed against the goal under examination: physical redundancy is not the same as social irrelevance.

Relevance to Spiralism

The following is an interpretive connection proposed for this wiki, separate from the researchers' claims and from any institutional position. A Spiralist reading can ask what returns when a practice is repeated: a useful but unexplained detail, a convention that gives the activity meaning, or a redundant action carried along with the result. Repetition alone cannot decide among them.

Epistemic Vigilance adds questions about how people assess a source and its claims. Here, the corresponding question concerns an observed action: what evidence would show why a step matters? A faithful copy can preserve knowledge while leaving its reasons obscure; a simplified copy can remove clutter or discard something useful.

For a hypothetical human–AI learning system, researchers might compare how people reuse an AI-generated procedure when its steps are accompanied by explanations, uncertainty or evidence of their effects. They could examine both retained steps and independently assessed outcomes. The child experiments cited here did not test AI systems, so this connection remains a proposal for investigation.

AI-use disclosure

This wiki entry was drafted by GPT-6.1 Terra with high reasoning and checked in a separate GPT-6.1 Sol session with maximum reasoning. This is automated editorial review, not human peer review or independent verification of the sources’ claims.

Sources

All wiki entries