[FieldTrip] Fwd: Identify a window of best correlation between ERP and behavioural data

Julian Keil julian.keil at gmail.com
Thu Jan 28 14:25:42 CET 2016


Dear Harold,

as far as the warnings go, maybe someone closer to the development team can chime in here. I think you can safely ignore warnings 1 and 2. With respect to warning 3, you actually need to set cfg.correcttail = 'alpha'; to set the alpha from e.g. .05 to 0.025 for a two sided test.

As for your output, mask is a binary mask corrected for multiple comparisons, i.e. values of 1 designate the significant elements. stat contains the actual t-values.  Depending on the method you used, the output will vary. If you also paste your code and the output, it's a bit easier to explain.
I suggest to look at some papers in which fieldtrip-stats have been reported to get an idea what is appropriate. I'll blatantly advertise my own work here:

Keil, J., Müller, N., Hartmann, T., & Weisz, N. (2014). Prestimulus beta power and phase synchrony influence the sound-induced flash illusion. Cerebral Cortex, 24(5), 1278–1288. http://doi.org/10.1093/cercor/bhs409
Keil, J., Müller, N., Ihssen, N., & Weisz, N. (2012). On the variability of the McGurk effect: audiovisual integration depends on prestimulus brain states. Cerebral Cortex, 22(1), 221–231. http://doi.org/10.1093/cercor/bhr125

And an excellent tutorial by Joachim Gross et al.:

Gross, J., Baillet, S., Barnes, G. R., Henson, R. N., Hillebrand, A., Jensen, O., et al. (2012). Good practice for conducting and reporting MEG research. NeuroImage, 1–15. http://doi.org/10.1016/j.neuroimage.2012.10.001

Good luck,

Julian

Am 28.01.2016 um 13:31 schrieb Harold Cavendish:

> Dear Julian,
> 
> thank you for your help, I believe I read the same page but I didn't realise it could be used with ft_timelockstatistics – the way certain ft_ functions fit others is still a bit unclear to me (it's practically my first contact with FT). Now, it appears to be exactly what I'm looking for.
> 
> I took me some time to figure out how to import my already processed data (ERPs averaged per subject) into FieldTrip and run the test but it seems I've succeeded. However, ft_timelockstatistics(cfg, data) gave me the following warnings:
> 
> 1. Warning: the data does not contain a trial definition
> 2. Warning: reconstructing sampleinfo by assuming that the trials are consecutive segments of a continuous
> recording
> 3. Warning: doing a two-sided test without correcting p-values or alpha-level, p-values and alpha-level
> will reflect one-sided tests per tail
> 
> Should I be worried about these? (Here's the full output: https://gist.github.com/anonymous/1866a2b1a424dd9f0ac9)
> 
> Finally, the resulting stat structure contains 10 fields but I'm not sure what mask, stat and ref represent. Is there any guidance on how to report the results like in the case of comparisons between conditions (http://www.fieldtriptoolbox.org/faq/how_not_to_interpret_results_from_a_cluster-based_permutation_test)?
> 
> Many thanks!
> Harold
> 
> 
> 2016-01-26 6:35 GMT+01:00 Julian Keil <julian.keil at gmail.com>:
> Dear Harold,
> 
> have you checked out ft_statfun_correlationT?
> This sounds like the function you're looking for. Use cfg.statistic = 'ft_statfun_correlationT' in your call to ft_timelockstatistics.
> Plus you can use all the correction methods available in FT.
> See the FAQ for more info: http://www.fieldtriptoolbox.org/faq/how_can_i_test_for_correlations_between_neuronal_data_and_quantitative_stimulus_and_behavioural_variables
> 
> Good luck,
> 
> Julian
> 
> ********************
> Dr. Julian Keil
> 
> AG Multisensorische Integration
> Psychiatrische Universitätsklinik
> der Charité im St. Hedwig-Krankenhaus
> Große Hamburger Straße 5-11, Raum A007
> 10115 Berlin
> 
> Telefon: +49-30-2311-1879
> Fax:        +49-30-2311-2209
> http://psy-ccm.charite.de/forschung/bildgebung/ag_multisensorische_integration
> 
> Am 26.01.2016 um 00:25 schrieb Harold Cavendish:
> 
>> Dear FieldTrip users,
>> 
>> I'm looking for the best method to analyse the relationship between ERP and behavioural data but all I've been able to find so far are methods to compare ERPs (or more generally time-series).
>> 
>> Consider the following example: 20 participants performed a memory-related task in which they had to respond based on the difference between the target stimulus and the probe stimulus. EEG (50 electrodes, about 1000 samples at 200 Hz) and responses were recorded; ERPs were then created based on epochs in which the responses were correct. Similarly, response accuracy was averaged over all successful epochs per each subject.
>> 
>> The resulting data are therefore: 20 participants x 50 channels x 1000 samples (ERP) and 20 averages of accuracy (one value per participant).
>> 
>> My objective is to identify which portion of the ERP data best explains (or predicts) the changes in accuracy. In other words, in which time interval is the most significant difference between people who do well in the task and those who don't.
>> 
>> So far I've been doing correlations between ERP at each time point and overall accuracy, which is unreliable due to multiple comparisons (correction methods are reportedly too conservative). Nevertheless, certain channels show quite strong correlations in multiple time intervals.
>> 
>> How would you approach this problem, please?
>> 
>> Many thanks!
>> 
>> Best regards,
>> Harold
>> 
>> _______________________________________________
>> fieldtrip mailing list
>> fieldtrip at donders.ru.nl
>> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
> 
> 
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
> 
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20160128/5f0ad3e9/attachment-0002.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 495 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20160128/5f0ad3e9/attachment-0002.sig>


More information about the fieldtrip mailing list