The feedback displaying issue

[mengcaopsy]

I was testing the bird experiment after adding the UI setting code. During the learning session, after I clicked a button, it switched to a feedback page with no options displaying on the screen. The first problem is that if my previous trial is correct, the feedback "You answer: undefined correct" will stay on the screen for the subsequent trial until I make a selection. The second problem with the feedback is that it says "Your answer: undefined incorrect. The correct answer is Jay." I'm not sure what the word "undefined" means. Since we don't have to show participants' responses, can we just keep the word "Correct/Incorrect. The correct answer is __"?

"uiSettings": { "displayReviewTimeoutAsBarOrText": "false", "displayReadyPromptTimeoutAsBarOrText": "false", "displayCardTimeoutAsBarOrText": false, "displayTimeOutDuringStudy": false, "singleLineFeedback" :true, "feedbackDisplayPosition" :"bottom", "stimuliPosition" :"left" },

What I think is good is below. If we can't do this, at least remove the additional words I mentioned above.

A minor point is whether we can make the background of the options to be squared. This isn't that important. I am just more used to the squared ones.

[imrryr]

Summary. @JRustyHaner

• "Your answer: undefined" seems to be just a bug, especially the way it is not removed for the next trial.
• is it possible "displayCorrectAnswerInCenter" is not working right, causing the gray box with the answer for feedback? It is what shows the gray box with the correct answer I think, and if false is default, it should have also been removed here. So is this a bug also.
• Can we control the fontsize of this feedback message, or just simply increase it a bit?

[imrryr]

Your answer undefined happens for me in some cases also

[imrryr]

It appears if I type a response, it works properly, with the colored word feedback, which is also missing. Note [timeout] should be after the response, not before, since it happened after

[JRustyHaner]

@imrryr could you be more specific please? what happened after?

[imrryr]

May we also get these details? Thanks for all the work! the critical changes despite the bugs.

• hide "your answer: X" on the correct feedback screen
• Make the score be right justified in a box that is less wide

[imrryr]

It currently says [timeout] undefined, but is should say the actual word they typed and [timeout] after that, not before.... it is not a timing issue it is the order of the two words.

so for instance, "Your answer: [timeout] undefined", it should be "Your answer: text entered [timeout]" timeout comes last if it occurs. @JRustyHaner

[JRustyHaner]

Thank you

[JRustyHaner]

@imrryr I had a default set wrong while testing.

The user's answer was displaying undefined because the original logic was looking for the value of a non existent text input. This has been fixed

score box is now right justified and smaller

font on the feedback will match the size in the tdf,

user answer is not displayed when correct

building now

[imrryr]

font on the feedback will match the size in the tdf, -- do you mean the same font as the stimulus?

[JRustyHaner]

the fontsize in the deliveryparms

[imrryr]

what are the range of values fontsize will accept? Only 1-5 or other values as well (for documentation) @JRustyHaner

[mengcaopsy]

The current feedback for the time-out is below. There are two timeouts and the feedback is far away from the image. Is there a way to move the feedback up a little bit?

The second bug Phil didn't mention in the summary is for the correct feedback. I used red circles and comment in the figure below.

[JRustyHaner]

@imrryr the size value corresponds with H1 through H6 tags in HTML

[imrryr]

@JRustyHaner in addition to the glitch with the correctness feedback persisting, and the timeout getting repeated twice and being a bit too close to the bottom, reported by Meng, I found the word "incorrect." was not capitalized. Also, I found that when I timedout, it did this for my case. It appears to be a simple error where it failed to produce the feedback due to an error in javascript.

[imrryr]

By the way, it is very easy to read the screen now compared to how it was. Much easier to focus on the key info; looking really good.

[JRustyHaner]

@imrryr building now

[JRustyHaner]

@imrryr deployed

[mengcaopsy]

The feedback is too long and goes to the second row. It may be good to make the words "Your answer: xx" smaller or make it an option that we can select to not include it.

[JRustyHaner]

@mengcaopsy the font size is controlled with the font size in your tdf. You may also remove it alltogether by using displayUserAnswerInFeedback: false.

[imrryr]

Yes, Rusty beat me to the reply. Thanks! He is correct.

You might also find that making "singleLineFeedback" :true, is a bad idea if you have that much feedback and want to retain displayUserAnswerInFeedback: true.

[JRustyHaner]

@imrryr @mengcaopsy although I found something else that broke, please bear with me

[JRustyHaner]

@imrryr @mengcaopsy I had a really silly typo in there, and if displayUserAnswerInFeedback is set to false, it will break. Rebuilding.

[mengcaopsy]

I see. Thank you for your work @JRustyHaner One more thing I am a little bit worried is about the options in the bird experiment. I have so many options. Even though it is now listed on the right side, the scroll bar appears. Each trial the screen automatically scrolls down a little bit. I am afraid if some participants use CTRL ++ to increase the page size, they can't see the options at once. I also found a gray box at the bottom of the page. Below shows the effect of increased page size.

[JRustyHaner]

@mengcaopsy that grey box is a placeholder for when feedback is given on the bottom and when the bar and text for the timeout appear. We can try to have that removed only for cases where its not being used, but should you turn on the timeouts, it has to be there.

The only solution I have for you for the buttons is to wrap them (to where they do not appear stacked on top of each other., but side by side). You run out of real estate having that many options. You may remove stacking by setting stackChoiceButtons: false.

[mengcaopsy]

@JRustyHaner If the button arrangement issue is solved, there is no need to fix the gray box. I will test what you said. Thanks!

[JRustyHaner]

@mengcaopsy please let me know if that works, that case hasn't been tested yet.

[JRustyHaner]

@mengcaopsy that wont fix it actually. I'm going to make the stackChoiceButtons a number, so you can specify how many columns you want

[mengcaopsy]

@JRustyHaner I have 10 options, so maybe 2 columns.

[JRustyHaner]

@mengcaopsy I just looked at it with 2, it looks great.

[mengcaopsy]

@JRustyHaner Is it correct to specify it as "stackChoiceButtons": "2"?

[JRustyHaner]

@imrryr @mengcaopsy we are changing the setting name, since that isn't a good description. Please remove stackChoiceButtons and add

choiceButtonCols: 2

with no quotes, since its a number

The spec will be choiceButtonCols: (number range 1-12, default 1) - specifies the number of columns of choice buttons to display.

[imrryr]

If you think that is best Meng. I don't understand why we need all the whitespace above the buttons. It seems like that is enough space if the button column widget were justified to the top a bit more. The button window seems to either shifted down in the screenshot. Also, the whitespace between buttons could be reduced a little it seems. Lets not complicate things though, but if it is easy to shift the box up a bit, please do so.

On Thu, Oct 26, 2023 at 10:40 AM Meng Cao @.***> wrote:

@JRustyHaner https://github.com/JRustyHaner Is it correct to specify it as "stackChoiceButtons": "2"?

— Reply to this email directly, view it on GitHub https://github.com/memphis-iis/mofacts-ies/issues/1296#issuecomment-1781377244, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADDLPK52RV4WNZIUFZ2VRSLYBKAATAVCNFSM6AAAAAA6P4AIDKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTOOBRGM3TOMRUGQ . You are receiving this because you were mentioned.Message ID: @.***>

-- Philip I. Pavlik Jr. @.*** http://optimallearning.org/

[mengcaopsy]

@imrryr @JRustyHaner If what Phil suggested is easy to fix, it makes sense to me. I'm not sure how the columns will influence participants' selection when the first column of options is close to the image.

[JRustyHaner]

choiceButtonCols: 2 is a bit less complicated, but I also have it set to remove the whitespace when the feedbackDisplayPosition: "bottom"

figured I'd do both

[JRustyHaner]

@imrryr @mengcaopsy building.

[imrryr]

Wait a sec I have another bug I am typing

[imrryr]

OK, it appears to not be respecting this "displayCorrectAnswerInCenter": true and also giving the message about what the correct items is on the top. When the correct answer is in center it should not also give it on top. it is one or the other. I think this is recurring and was correct before.

[JRustyHaner]

Probably because of that typo I just fixed. I'll look at it again in a minute. I've got a tree guy coming to the house

[imrryr]

No worries.

[mengcaopsy]

Since the scroll bar is still there, can we narrow down the spacing between the options? Is it easy to fix? Or maybe the word MoFaCTS on the top left shouldn't occupy so much space on the top.

[imrryr]

I don't think the build was deployed yet @mengcaopsy

[JRustyHaner]

@imrryr @mengcaopsy building again, fixing @imrryr most recent comment

[mengcaopsy]

@JRustyHaner How about the option arrangement issue I just mentioned? Is there a way to fix it?

[imrryr]

I'm I'm not sure I understand your last comment. The version you were showing in your last comment was not the latest version.

On Thu, Oct 26, 2023, 1:20 PM Meng Cao @.***> wrote:

@JRustyHaner https://github.com/JRustyHaner How about the option arrangement issue I just mentioned? Is there a way to fix it?

— Reply to this email directly, view it on GitHub https://github.com/memphis-iis/mofacts-ies/issues/1296#issuecomment-1781622023, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADDLPK4EVBG4PJFDQMP76GTYBKSVXAVCNFSM6AAAAAA6P4AIDKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTOOBRGYZDEMBSGM . You are receiving this because you were mentioned.Message ID: @.***>

[mengcaopsy]

I see. Now it is fixed. Thank you! @JRustyHaner @imrryr But the feedback seems farther away from the image now.

[JRustyHaner]

@imrryr @mengcaopsy the most recent build was just deployed, and my neighbor can now see out of her driveway. @imrryr The sidewalk near my house should be almost walkable now

[JRustyHaner]

@mengcaopsy It will be, it will always appear at the bottom of the screen. You may use a larger image, but it won't blow up the image for you since it will pixelate it and make it look bad.

[imrryr]

While the problem wasn't very big problem Rusty it was that the feedback was too close to the bottom of the screen so that the saccade of the learner is much further away. Actually now that I think about it in this feedback mode it would be very nice to put the feedback inside the gray box which is much higher on the screen. However I think we would be happy with any solution that just brought them closer together the stimulus and the feedback basically that spot where the correct answer is put in the gray box is completely unneeded in this mode when you produce the correct answer in the feedback statement. @Rusty Haner @.***>

On Thu, Oct 26, 2023, 2:19 PM JRustyHaner @.***> wrote:

@mengcaopsy https://github.com/mengcaopsy It will be, it will always appear at the bottom of the screen. You may use a larger image, but it won't blow up the image for you since it will pixelate it and make it look bad.

— Reply to this email directly, view it on GitHub https://github.com/memphis-iis/mofacts-ies/issues/1296#issuecomment-1781760992, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADDLPKYWVPXUVSGUD2JJYCDYBKZVDAVCNFSM6AAAAAA6P4AIDKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTOOBRG43DAOJZGI . You are receiving this because you were mentioned.Message ID: @.***>

[JRustyHaner]

@imrryr the problem, of course is when the stimulus is larger than the available screen size. If we don't anchor the feedback to the bottom of the screen, then stimuli that are larger than the screen size will push the bottom feedback to where you have to scroll down to see it.

[imrryr]

Stimuli are either scaled or smaller than the size wher they need scaling. Mofacts is not intended to support stimuli larger than a screen. It shouldn't be an issue.

On Thu, Oct 26, 2023, 2:37 PM JRustyHaner @.***> wrote:

@imrryr https://github.com/imrryr the problem, of course is when the stimulus is larger than the available screen size. If we don't anchor the response to the bottom of the screen, then stimuli that are larger than the screen size will push the bottom feedback to where you have to scroll down to see it.

— Reply to this email directly, view it on GitHub https://github.com/memphis-iis/mofacts-ies/issues/1296#issuecomment-1781785214, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADDLPK7R7KBFL3GMTBZWFO3YBK3XVAVCNFSM6AAAAAA6P4AIDKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTOOBRG44DKMRRGQ . You are receiving this because you were mentioned.Message ID: @.***>