Incoherent error message from QuestHandler in PsychoJS

URL of experiment:

Description of the problem:
We ran a Quest simulation, just to make sure we’re doing everything right. We were surprised to get this “error” message:

run.pavlovia.org says
QuestUpdate: intensity -2.1941879156398305 out of range
-2.1989700043360187 to 2.8010299956639813. Pdf will be inexact.
Suggest that you increase “range” in call to QuestCreate.

The opening statement is false. In fact, the number -2.1941879156398305 is in the range -2.1989700043360187 to 2.8010299956639813.

We haven’t explored enough to decide whether this is a false alarm, or a garbled report of a real problem.

H Denis,

Thanks for reporting this issue. We think it’s a bug in the jsQUEST library, so I took the liberty of reporting an issue in the jsQUEST GitHub repo.

Best, Thomas

Thank you for the valuable information.

I am one of the developers and will find the cause as soon as possible.

It seems that the URL of your experiment is blank.
Could you please show me the URL or source code of the program where you saw this error?

I have probably found the cause and fixed the bug.

I have run the following code with jsQUEST v1.0.1 and confirmed that I get the same error.

const tGuess = -1
const tGuessSd = 2
const pThreshold = 0.82;
const beta = 3.5;
const delta = 0.01;
const gamma = 0.5;

let q = jsQUEST.QuestCreate(tGuess, tGuessSd, pThreshold, beta, delta, gamma);
tTest = jsQUEST.QuestQuantile(q);
q = jsQUEST.QuestUpdate(q, -3.499999, 1);

The error message is

QuestUpdate: intensity -3.499999 out of range -3.5 to 1.5. Pdf will be inexact. Suggest that you increase “range” in call to QuestCreate.

I have released jsQUEST v1.1, which fixes this bug.

Here is a brief explanation of the cause.

I ported Prof. Pelli’s MATLAB code to JavaScript; in MATLAB, array indices start at 1, but in JavaScript, they start at 0. This seemed to be the cause of the problem.

I am very grateful to Prof. Pelli for his important remarks and the MATLAB code.

1 Like

Dear kurokida
Thanks for your kind remarks, and for fixing the bug. That’s great.
Best
Denis

photo

Denis Pelli
Professor of Psychology & Neural Science, New York University

+1-646-258-7524 | denis.pelli@nyu.edu

http://psych.nyu.edu/pelli/ | Skype: denispelli

http://denispelli.com

1 Like