Intrinsic motivation

Reinforcement learning, Robotics

references :

According to (Ryan, Deci 2000) (pp. 56),

Intrinsic motivation is defined as the doing of an activity for its inherent satisfaction rather than for some separable consequence. When intrinsically motivated, a person is moved to act for the fun or challenge entailed rather than because of external products, pressures, or rewards.

It is defined by contrast with extrinsic motivation

Extrinsic motivation is a construct that pertains whenever an activity is done in order to attain some separable outcome. Extrinsic motivation thus contrasts with intrinsic motivation, which refers to doing an activity simply for the enjoyment of the activity itself, rather than its instrumental value.

Intrinsic motivation (IM) is different from internal motivation (as opposed to external motivation).

For example, an autonomous system learning to walk with a reinforcement learning algorithm could get its reward from a camera mounted in the room measuring the distance traveled (external) or from its own sensors measuring an analogous quantity (internal). However both these setups correspond to extrinsic rewards.

The problem with the concept of intrinsic reward is that its definition relies on nebulous concepts such as the “fun”, the “challenge” or the “novelty” of a goal. Plenty of contradictions may arise when looking for specific examples of IM. Which are partly due to science’s limited understanding of human or animal motivation.


  1. . . "Intrinsic and Extrinsic Motivations: Classic Definitions and New Directions". Contemporary Educational Psychology 25 (1):54–67. DOI.


← Back to Notes