Description
Subtract falseAlarmProbit from hitProbit for dPrime and return minus half their sum as criterion.
Signal-detection measures separate discriminability from a response criterion under an explicit observation model. Standard-normal quantiles are transformed inputs; providing those quantiles is different from supplying raw hit and false-alarm rates.
Subtract falseAlarmProbit from hitProbit for dPrime and return minus half their sum as criterion.
When to use this tool
Use Signal Detection Dprime One Probit Gap when you need the final d-prime and criterion calculation from supplied probits.
How to use the tool and how it works
- Enter hitProbit, falseAlarmProbit. Use the units and conventions below.
- Run the tool. Correct any format, range, or cross-field input error before interpreting the output.
- Read dPrime, criterion. Use the output convention below.
Subtract falseAlarmProbit from hitProbit for dPrime and return minus half their sum as criterion.1
The equal-variance normal model defines d′ as the hit/false-alarm quantile difference and criterion as minus half their sum. This tool starts from supplied probits.2
| Input field | Meaning and units | Accepted input |
|---|---|---|
| hitProbit | Already transformed standard-normal quantile of hit rate. | Required. Finite number. |
| falseAlarmProbit | Already transformed standard-normal quantile of false-alarm rate. | Required. Finite number. |
| Output field | Interpretation | Returned shape |
|---|---|---|
| dPrime | D Prime | Required. Finite number. |
| criterion | Criterion | Required. Finite number. |
{
"input": {
"hitProbit": 1,
"falseAlarmProbit": -1
},
"output": {
"dPrime": 2,
"criterion": 0
}
}Limitations and assumptions
- Inputs are already standard-normal quantiles, not raw probabilities or percent. No inverse-normal transform, extreme-rate correction, diagnostic interpretation or equal-variance model fit occurs.
- Inputs must satisfy both the field types and the cross-field conditions above. Invalid input causes an error; a successful false flag or documented placeholder is a different result.
- Numeric calculations use binary floating point unless exact decimal-integer arithmetic is explicitly stated. No hidden measurement tolerance or decimal rounding is added; nonfinite or out-of-contract results are rejected. Numeric signed-zero results are exported as zero.
Alternative or Complementary methods
Use an appropriate signal-detection analysis for raw counts or rates. Handle extreme rates, task design, distribution assumptions and uncertainty before interpreting discriminability.
References
-
Calculation and input contract: experimental-psychology/signal-detection-dprime-one-probit-gap — tool.io
-
Calculation of signal detection theory measures — Stanislaw and Todorov, Behavior Research Methods, Instruments, & Computers
Similar or alternative tools
- neuroscience/signal-detection-from-counts
Signal-detection sensitivity separates hits from false alarms through normal quantiles.
- neuroscience/signal-detection-criterion
In equal-variance Gaussian signal detection, criterion c describes response tendency using hit and false-alarm fractions.