The P300 is an event-related potential (ERP) component — a reliable, time-locked positive deflection in the scalp EEG signal occurring approximately 300 milliseconds after an infrequent or task-relevant stimulus. It was first described by Samuel Sutton and colleagues at Columbia University in 1965.
Neural Origin
The P300 reflects cognitive processing of rare or meaningful events. It is maximal at central-parietal scalp electrodes (Cz, Pz) and is generated by coordinated neural activity in prefrontal cortex, parietal cortex, and hippocampal areas. The P300 is not specific to visual stimuli — it occurs in response to any rare or target stimulus in any sensory modality (visual, auditory, or tactile).
Two sub-components are distinguished:
- P3a (early, ~280ms): Frontal P300 associated with involuntary orienting to novel stimuli
- P3b (late, ~350-500ms): Parietal P300 associated with voluntary target detection
P300 in BCI: The Oddball Paradigm
The P300 BCI was first demonstrated by Farwell and Donchin in 1988. In the classic P300 Speller:
- A matrix of letters (6x6 = 36 characters) is displayed on screen
- Rows and columns flash in a random order ("oddball" paradigm)
- When the row or column containing the user's target letter flashes, the user's brain registers the target event, generating a P300
- A classifier identifies which row AND which column generated P300 responses, identifying the target character
- The system then "types" that character
Performance and Limitations
P300 BCIs are non-invasive and require no training of motor imagery. However:
- Speed: Typically 5-10 characters per minute — much slower than motor imagery or intracortical BCIs
- Fatigue: Requires sustained attention; performance degrades with fatigue
- In the clinic: Useful for patients with ALS who retain attention and visual function
- Cognixion ONE: Uses P300 detection combined with SSVEP and eye tracking in an AR interface, improving speed over classic P300 spellers by integrating multiple BCI paradigms
P300 vs. Other ERP Components
| Component | Latency | BCI Use | |---|---|---| | N200 | ~200ms | Error detection, mental workload | | P300 | ~300ms | Target detection, spellers | | N400 | ~400ms | Language processing | | Error-related negativity (ERN) | ~80ms post-error | Closed-loop error correction |