Do Ravens Have Theory of Mind?

What "Theory of Mind" Means

[science] Theory of mind (ToM) is the ability to attribute mental states — beliefs, desires, intentions, knowledge — to others and use those attributions to predict and explain behaviour. It develops in human children around age 3–4.

In animal cognition research, ToM is usually tested in components:

• Gaze following: Does the animal follow another's line of sight?
• Knowledge attribution: Does the animal distinguish between a knowledgeable and an ignorant competitor?
• False belief understanding: Can the animal predict what someone will do based on a belief that is wrong?
• Perspective-taking: Does the animal understand what another can see from their vantage point?

Ravens have been tested on several of these components with positive results.¹

Key Studies

[science]

• Bugnyar & Heinrich (2005): Ravens differentiating knowledgeable from ignorant competitors in food caching. A raven that cached food while watched by a dominant competitor would protect the cache differently than when watched by a bird that was absent (ignorant). This suggests the cacher distinguishes the competitor's knowledge state — a component of ToM. Proceedings of the Royal Society B, 272, 1641–1646.
• Bugnyar et al. (2016): The "peephole paradigm" — a raven was able to see a competitor through a small hole, but was tested on its behaviour as if the competitor might have seen it, without direct gaze exchange. Ravens adjusted cache-protection behaviour in response to the mere possibility of being observed. This is a more demanding test. Nature Communications, 7, 10506. DOI: 10.1038/ncomms10506
• Pika & Bugnyar (2011): Referential gestures in ravens — ravens used pointing-like gestures toward objects while monitoring a partner's gaze, suggesting awareness of attentional states. Current Biology, 21(20).

Gaze Following

[science] Ravens follow the gaze of conspecifics and humans — they look in the direction another individual is looking, particularly around barriers. Bugnyar et al. (2004) documented gaze following in ravens, including geometrically correct gaze following around obstacles (looking around a barrier to see what a conspecific is looking at, not just looking in the same direction). This component of perspective-taking is well-established in ravens.²

What Is and Is Not Established

[science] Honest summary of the evidence:

• Well-established: Ravens follow gaze; ravens distinguish knowledgeable from ignorant competitors; ravens adjust cache-protection behaviour in response to observation
• Demonstrated but contested: Whether the peephole paradigm (Bugnyar 2016) demonstrates genuine perspective-taking or a learned rule about "holes → might be watched"
• Not yet demonstrated: Full false-belief understanding (Standard ToM test); ravens have not been shown to predict behaviour based on beliefs they know to be false
• Contested interpretation: Whether evidence for knowledge attribution reflects genuine mentalising (representing others' mental states) or sophisticated associative learning that mimics mentalising without it

The honest scientific position is that ravens show some components of ToM to a degree not seen in most animals — but whether this constitutes "true" theory of mind in the philosophical sense remains actively debated.^1,3