Confidence scale

Brooke90

Hello Everyone,

I have set up a basic facial recognition experiment for my dissertation but I have decided to add a confidence scale on a sketchpad and keyboard response (indicating confidence by pressing keys 1-10) However, when I am importing data in to SPSS its coming up as NA?

Any idea why its not logging the keyboard presses?

Thanks,

Brooke

Daniel

Hi Brooke,

I think we need a little more information to go on to pinpoint the problem. Could you post a screenshot of the experiment overview section which contains the stimulus display (confidence scale) and keyboard response item?

My initial guess is that you did add a keyboard_response item, but did not add a logger item afterwards, causing the keyboard responses not to be logged, but I can only say for sure if I can see what you have done.

Brooke90

Hi Daniel,

Thank you for your reply, Initially I didn't have a logger then noticed upon data collection that I didn't have one so I fixed that problem, however, it still doesn't log the response? I added a Trial sequence before the confidence scale sketchpad and then keyboard response as well but it just seem to log the response at all?

screenshot as per below, I would greatly appreciate the advice!

Thank you Brooke

eduard

Hi Brooke,

I can't see a logger in the screenshot. So, I suppose you added it only to the end of the experiment. However, what you need to do is adding it after every time you expect a response from the participant (normally in the end of every trial sequence). Does that make sense?

Eduard

Brooke90

Hi Eduard,

The first block of my experiment is purely showing pictures while the second block I have added a logger as the participants are required to press 'z' or 'm' if they have seen a face from block 1. This seems to be okay upon checking the data file afterward, however, I wish to gain a key response of 1-10 to determine how confident the participants were in their responses, to do this I have added a new sequence, the sketchpad with the question (set to 0) a keyboard response with [response] and then the logger? does this seem to be the right way to gain responses for a confidence scale?

The thing that is confusing me is that I have added a sketchpad and keyboard response at the beginning of my experiment to get participants age and gender by pressing 1 or 2 and this works fine without a logger so I don't understand why the confidence scale isn't working with or without a logger?

eduard

The logger writes the current values of all the variables that are currently stored to a file. Age and gender you only ask once in the beginning, so their values don't change and you'll see them appear in the logfile after the first logger is called.

The specific why you don't see the other variables is a little hard to tell without seeing the experiment in its whole, specifically the loops with the rating scale. So, if you could attach the experiment, I might have a better idea how to help.

Eduard

Brooke90

Facial_recognition.osexp

You are a star Eduard, thank you so, so much for taking the time to look at it, I really am at a loss after thinking everything was okay to start recruiting participants and now I am way behind with data collection/analysis and writing up the report, all to be completed by 29th of April.

Thank you again!!

Brooke :)

eduard

Hi Brooke,

The last sequence is not necessary, you can also just call the three items in it directly. The problem with the experiment is that you specified a correct response ([response]). Why is that? This variable is not defined at that stage. And it is also a bit weird to prespecify a correct response for a confidence judgement, no? If you leave this free empty, the experiment will finish properly.

Eduard

(btw, there were a few weird things with the experiment, e.g. uneven ITIs, I don't know whether this is because of my computer, but perhaps check it before you properly start with data collection)

Brooke90

Hi Eduard,

Thanks again for the reply, the experiment is a template from my university so I don't understand what you mean by the weird things? I was advised my tutor to add a confidence scale at the end in order to have a second dependent variable for my project, again advice from university was to use this in the form of a sketch pad which is why I done it this way, I have used correct response so it records the participants correct response (either Z or M) as I am calculating number of correct facial matches as my other dependent variable. However as you can see its not recording the data.

what do you mean by uneven ITIs?

Thanks again

Brooke

eduard

I have used correct response so it records the participants correct response (either Z or M)

That makes sense for the actual task, but not for the confidence rating. So, they keyboard response for the confidence rating does not need a correct response

what do you mean by uneven ITIs?

Time between two successive images

so I don't understand what you mean by the weird things?

If you don't see weird things, then never mind. I don't know the experiment and how it is supposed to look, so maybe everyhthing was fine, but just different from what I expected.

Eduard