Scientists have decoded the melodies people imagine in their heads by reading electrical signals directly from the surface of the brain, then reconstructing the songs note by note.
The Research
In a study led by Jii Kwon and Chun Kee Chung at Seoul National University, published in the journal eNeuro (Society for Neuroscience), researchers worked with 10 epilepsy patients who already had electrodes placed on the surface of their brains for clinical seizure monitoring. This intracranial setup gave the team a rare, high-resolution window into real-time brain activity.
The experiment unfolded in three stages. First, participants quietly listened to the opening bars of familiar children's songs. Next, they silently imagined how each melody continued, filling in the following bars in their minds. Finally, they hummed their imagined tunes aloud so researchers could verify what they had pictured.
The key finding: brain activity carried enough spectral and temporal information to decode the relative pitches of the imagined notes. Critically, the model targeted relative pitch classes — do, re, mi, fa, sol, la — rather than absolute frequencies. As Kwon explains, "This is useful because people often recognize melodies by the relationships between notes, even when the same melody is played in a different key."
By aggregating note-level predictions across time, the team reconstructed melodic contours that matched the structure of the songs participants had imagined. The sample of 10 patients is small, as is typical for invasive intracranial research, but the within-subject data were rich — hundreds of electrodes recording millisecond-by-millisecond.
Why It Matters
This is direct neural evidence that imagined music lives in measurable, structured brain activity — not as vague "mental noise" but as decodable pitch sequences. That has two big implications. First, it deepens our understanding of auditory imagery, the same cognitive machinery you use when a song gets stuck in your head or when you "hear" a melody just by reading its title. Second, it's a proof of concept for music- and speech-based brain-computer interfaces that could one day help people with severe motor or speech impairments, such as Amyotrophic Lateral Sclerosis, express what they're thinking.
For the rest of us, the study is a reminder that imagination is not a soft, unmeasurable thing. It runs on specific neural code — and that code can be trained.
What You Can Do
- Practice melodic imagery. Listen to a familiar song, pause it, and try to "hear" the next few bars in your head before checking. This exercises the same pitch-class representations the study decoded.
- Transpose in your mind. Hum a tune, then hum it again starting on a different note. You're training relative pitch — the cognitive anchor of melody recognition.
- Test your auditory working memory. Try repeating back short sequences of tones or notes; this overlaps with the temporal tracking used in the decoding model.
- Keep your ears and mind active. Musical training and active listening are associated with stronger auditory imagery and working memory in healthy adults.
Source: Neuroscience News
Curious about your own brain? Take our free adaptive IQ test or try 306 brain training levels.