Externalization is the percept that a sound comes from outside the head, not stuck inside the ear canal

What it is

Listening to music through headphones, many people feel the sound is "playing inside their head" rather than arriving from some direction in external space. By contrast, the same content can sometimes sound as if it's coming from a specific location in the room, as though a speaker were really sitting there. That percept of "the source is outside the skull" is called externalization. It's one of the core goals of spatial audio, and it's a related but distinct thing from "can direction be judged correctly."

Externalization is easy to conflate with "accurate direction judgment." In fact a sound whose direction is judged perfectly correctly can still be perceived as stuck inside the head — judging "this sound leans left" and perceiving "this sound is coming from a spot a meter to my left" are separable perceptual dimensions.

Why it happens

The brain's decision about whether a sound originates in external space or is "trapped" inside the head doesn't rest on a single cue — it rests on whether a whole set of evidence jointly points to "this is a sound that traveled through real space and was naturally filtered by the body": appropriate directional cues (interaural time difference, interaural level difference, pinna spectral filtering), plus whether the sound carries the distance and environmental cues a real space would produce (such as reverberant content). When this evidence is sufficient and mutually consistent, the brain leans toward judging the source as an external entity; when a sound lacks the traces such natural filtering would leave — for example, the same signal sent identically to both ears — the brain finds no evidence supporting "this passed through external space," and falls back on the simplest default explanation: the sound originated inside the head.

Externalization is therefore an inferential outcome, not something decided by any single cue alone: directional cues can be complete while distance and environmental cues are absent, still yielding "direction correct but not externalized" — which is exactly why externalization and pure direction judgment are independent of each other.

Studying it

The standard approach measures direction-judgment accuracy while separately collecting a participant's subjective report of "does this sound feel like it's outside or inside your head," usually on a continuous scale (fully inside to fully outside), recording direction accuracy and externalization rating as two independent dependent variables rather than assuming they always move together.

Common independent variables: presentation method (direct headphone playback, with head-related transfer function processing added, with simulated reverberation added), whether the processing is individualized. Common dependent variables: direction judgment error, externalization rating.

Where it stops holding

• Externalization and direction-judgment accuracy are separable dimensions; an experiment that measures only one of the two risks a lopsided conclusion about whether a spatial audio setup is "good."
• Externalization rating can be affected by a listener's familiarity and adaptation to a given presentation method — the first exposure to a spatial audio format may show lower externalization, improving with accumulated listening experience, so a first-exposure rating alone shouldn't be treated as conclusive.
• This entry covers what the externalization percept is and what determines it in general; which specific presentation techniques tend to fail, and which cues are decisive, are developed separately in the other entries in this group.

Applying it

• When evaluating a spatial audio setup, don't accept it based only on "is direction correct" — design a separate step for listeners to report whether the sound is perceived as coming from outside the head, recording the two metrics separately.
• If the product goal is an "immersive" sense of space, externalization itself is a directly measurable, directly optimizable target — there's no need to wait until direction judgment fails to discover that spatial presence is missing.
• Verification: have listeners rate both direction accuracy and externalization for several candidate spatial audio presentation options under blind conditions, recording the two dimensions separately, rather than substituting a vague overall impression of "does it feel spatial" for two independent metrics.

Related

• Same group: A3.14.2 Ordinary stereo headphones struggle to externalize, prone to producing an in-head localization illusion · A3.14.3 The head-related transfer function is individual, and generic parameters reduce localization accuracy · A3.14.4 Reverberation ratio is the primary cue for judging source distance, and directional cues cannot substitute for it · A3.14.5 Binaural recording reconstructs natural spatial cues better than plain stereo
• Nearby: A3.05 Sound source localization
• Search terms: externalization · in-head localization · binaural rendering · spatial audio