ITR Atlas
← All interfaces

High-Speed SSVEP Speller (Chen et al., 2015)

Measured by Chen, Wang, Gao, Jung & Gao · PNAS 112(44) (2015)

EEG SSVEP 2015

Inputs

The measured or assumed values behind the calculations, each with its source.

N = 40
40-character speller; each target tagged with a 0.5 s joint frequency–phase-modulated flicker
ITR = 4.45 bits/s
Mean online ITR across subjects, including the 0.5 s gaze-shift time (Abstract). Peak individual 5.32 bits/s; spelling rate up to ~60 char/min (~12 wpm).

Strictest ITR

Each scoring method is an upper bound on the channel, so the headline is the strictest (smallest) one for this entry. Use the score selector on the home page to view any single method across entries.

Strictest Shannon (text) Recomputed
Character-entropy throughput (realized text entry)
Net of English redundancy
0.9 bits/s

  1. Correct characters per minute

    ~1 selection/s (0.5 s flicker + 0.5 s gaze shift) → ~60 char/min × 89.83% ≈ 54 correct char/min

    Each selection emits one character; this is the rate of correct English text produced (~12 wpm gross).

  2. Bits per character

    H(English) ≈ 1.0 bit/char (Shannon), the same predictor used for QWERTY and the other text entries

  3. Information transfer rate

    54 char/min × 1.0 bit/char ÷ 60 s/min = 0.9 bits/s

What counts as a bit depends on the action space. The number of distinguishable actions and how likely each one is are design choices of the task, not the sensing hardware. The same modality can present a fixed set of targets, a set pruned per step by a grammar or language model, or a continuous control space. Each of these changes how many actions are live and how the probability mass is spread, and therefore the information per selection. Read the action space below before comparing headline numbers across entries.

Action space

What the user can produce at each step, and how those options are distributed.

Structure
Fixed set of targets
Size
40 distinguishable actions
Prior
Context-conditioned: likelihoods depend on prior actions
Notes
40 flicker-coded characters; the decoder classifies which flicker frequency the user is gazing at: a classifier, not Fitts pointing (selection time is set by the flicker-integration window, not by target distance/size). The realized output is English text, so the reference uses character-entropy (~1 bit/char) like the other text entries; the Wolpaw-over-40 figure counts log2(N) per selection and is kept as a secondary classifier metric. The c-VEP entry (EEG2Code) is a useful comparison point because it reports very large code-space discrimination separately from active spelling throughput.

Comparability The strictest bound here is the Shannon entropy of the output text, under one predictor held constant across the whole atlas (≈1 bit per character). That shared predictor makes it directly comparable to every other text entry (keyboards, spellers, silent speech and speech BCIs) regardless of prior or vocabulary size. For most text interfaces it comes out tighter than the raw-selection bounds, but not always. Where a small vocabulary makes Wolpaw tighter, that wins instead. Any Fitts, Wolpaw or log₂(N) figure shown below is another bound on the same channel. Switch the home-page score selector to compare one across entries.

Other score types

Bounds the atlas keeps out of the default strictest headline: as-reported figures, alternate task conditions, or raw-channel ceilings that shouldn't win the headline by default. Each still carries a score type, so the home-page selector ranks this entry on it when you choose that type. Read its derivation before comparing across entries.

Wolpaw Author-reported · reproduced
Wolpaw bitrate over N = 40 targets (authors' reported ITR)
Uniform 1-of-40 classifier metric, shown for comparison
4.45 bits/s

  1. Bits per selection (Wolpaw, N = 40 at 89.83%)

    B = log2(40) + 0.8983·log2(0.8983) + 0.1017·log2(0.1017/39) ≈ 4.31 bits/selection

    Forward from the online accuracy (89.83%) over the 40 targets. Perfect-accuracy ceiling is log2(40) = 5.32 bits, approached by the best subject.

  2. Authors' reported online ITR (includes the gaze-shift time)

    ITR = 4.45 bits/s  (mean across subjects; 0.5 s flicker + 0.5 s gaze shift ≈ 1 selection/s)

    Author-reported and verified: B ≈ 4.31 bits/selection at ≈1 selection/s reproduces it. This counts log2(N) per selection, the classifier metric, not the 1 bit/char the atlas-ranked text figure holds every text entry to.

Nuyujukian Recomputed
Nuyujukian achieved bitrate over N = 40 targets
Achieved-bitrate view of the speller, shown for comparison
4.21 bits/s

  1. Achieved-bitrate credit per net-correct selection

    N = 40 → log2(N − 1) = log2(39) = 5.29 bits per net-correct selection (field-standard achieved bitrate, e.g. Webgrid; Nuyujukian 2015, which introduced the metric, used log2(N)).

  2. Net-correct selection rate

    net-correct = 2P − 1 = 2(0.8983) − 1 = 0.797 of selections. At ~1 selection/s (0.5 s flicker + 0.5 s gaze shift) → 0.797 correct/s.

    A speller error commits the wrong character rather than timing out, so incorrect = 1 − P. Same N (40), online accuracy (89.83%) and selection rate (~1/s) as the entry's Wolpaw calc. Netting each wrong selection against a correct one (2P − 1) drops the achieved figure a little below the 4.45 bits/s Wolpaw ITR; the gap widens as accuracy falls.

  3. Achieved bitrate

    5.29 bits × 0.797 correct/s = 4.21 bits/s.

Source

Authors
Chen, Wang, Gao, Jung & Gao
Publication
PNAS 112(44), 2015
Paper
10.1073/pnas.1508080112