tag:blogger.com,1999:blog-1405968300258104460.post5606402792799477815..comments2024-03-27T01:18:24.341-04:00Comments on Andy's Brain Blog: Writing Out Timing Files from E-PrimeAndrew Jahnhttp://www.blogger.com/profile/16435706598096921650noreply@blogger.comBlogger6125tag:blogger.com,1999:blog-1405968300258104460.post-19755159817324136932017-02-06T02:54:01.572-05:002017-02-06T02:54:01.572-05:00All right, many thanks again !
All the best
Pier...All right, many thanks again !<br /><br />All the best<br /><br />Pierre-YvesAnonymoushttps://www.blogger.com/profile/03916153025900437276noreply@blogger.comtag:blogger.com,1999:blog-1405968300258104460.post-40720841348235828982017-02-03T13:21:32.011-05:002017-02-03T13:21:32.011-05:00Yes, that's right. If there were some kind of ...Yes, that's right. If there were some kind of drift in the timing errors, that might be cause for concern, but I would just run the experiment, write out the timing files, and see whether there is a significant drift over time.<br /><br />-AndyAndrew Jahnhttps://www.blogger.com/profile/16435706598096921650noreply@blogger.comtag:blogger.com,1999:blog-1405968300258104460.post-53267021560679765642017-02-02T15:59:55.105-05:002017-02-02T15:59:55.105-05:00Hi Andy,
Thank you so much for your kind & pr...Hi Andy,<br /><br />Thank you so much for your kind & prompt answer, and for the great job you share on your blog.<br />Ok, I feel quite confident now following your post, and I guess the reason why your code does not take into account the onset delays is also included in your answer. Meaning, onset errors inferior than 100 ms won't affect the slow HRF we aim at sampling, right ?<br />I was also afraid that in some way, either with cumulative or event mode, the timing error may cumulate across a run, but this is not the case.<br />All the best & thanks again,<br /><br />Pierre-YvesAnonymoushttps://www.blogger.com/profile/03916153025900437276noreply@blogger.comtag:blogger.com,1999:blog-1405968300258104460.post-21561691714083589512017-02-02T14:20:28.121-05:002017-02-02T14:20:28.121-05:00Hi Pierre-Yves,
It depends on what you want. Cumu...Hi Pierre-Yves,<br /><br />It depends on what you want. Cumulative timing will give you more accurate timing across the entire experiment - e.g., if you have a 5-minute block, it will stop at 5 minutes - whereas event timing will ensure that each object is presented for the duration that you specify. If there are any delays (for example, due to the computer processing other things in the background), cumulative timing will subtract that delay from the time the next object is to be presented, so that the total time across the experiment remains constant.<br /><br />In the past I've used event timing, and I've found that the errors in stimulus presentation were no more than a couple of hundredths of a second - not a problem for most FMRI research, given the slow BOLD response that we typically measure. Also, there may be cases where taking time from another slide can affect the participant's experience; for example, if you're presenting words in rapid serial format, say at 300ms per word, and any reduction could lead to comprehension confounds. (This is a hypothetical example; I'm not actually sure whether a difference that small could cause any confounds, but you get the point.)<br /><br />These days I recommend using cumulative timing, except in cases where you absolutely need the object to be presented for the full duration. Either way, this won't affect the code I gave you. Pre-release will, but that's another matter.<br /><br /><br />Best,<br /><br />-AndyAndrew Jahnhttps://www.blogger.com/profile/16435706598096921650noreply@blogger.comtag:blogger.com,1999:blog-1405968300258104460.post-7307502829231700472017-02-01T04:33:50.941-05:002017-02-01T04:33:50.941-05:00Hi Andy
Many thanks for posting such a useful tric...Hi Andy<br />Many thanks for posting such a useful trick ! I have a question regarding the timing mode you would recommend when using EPrime: do you think "Cumulative" timing should be preferred and if so, could the use of the "Event" timing mode affect your code in some way ?<br />Thanks by advance!<br />Pierre-YvesAnonymoushttps://www.blogger.com/profile/03916153025900437276noreply@blogger.comtag:blogger.com,1999:blog-1405968300258104460.post-87512173359467410032015-12-03T12:49:56.855-05:002015-12-03T12:49:56.855-05:00Hi Andy, thanks so much for posting this - it prom...Hi Andy, thanks so much for posting this - it promises to be really helpful! Would you mind posting links to the E-Studio files (or to a dummy file which writes out data for a single condition)? I'm new to using Eprime's scripting capabilities & being able to see when you define & call variables would be extremely helpful! Thanks again,<br /><br />JohnAnonymoushttps://www.blogger.com/profile/03924611838306603179noreply@blogger.com