[Erppcatoolkit-support] EP Toolkit 2.53 released

[Joseph D. <jd...@me...>]

EP Toolkit version 2.53 has been released:
https://sourceforge.net/projects/erppcatoolkit/ <https://sourceforge.net/projects/erppcatoolkit/>

Aside from the usual bug fixes, I’ve been focusing on further improving support for spectral analyses and an improved Scan function for examining and editing continuous data files.

I would like to note that while a recent methodology paper (Clayson, Baldwin, & Larson, 2013) reported that my centroid latency measure (an option in the Window function) is "biased", the authors generously allowed me to examine the script that ran their simulation analyses and I found that contrary to their methods description they had not randomized the phase of the background sine waves (simulated noise), resulting in a confound.  So in fact the "bias” in the centroid measure was actually due to it to being more sensitive (correctly) to the presence of the confounded noise than the competing methods (which explains why this "bias" was stronger the larger the noise-to-signal ratio).  The authors have accepted this critique and responded that they plan to submit an erratum to their paper.

I would also like to note that I am currently in a soft-money position so if you have found the EP Toolkit to be useful for your research or think that I could otherwise contribute to your studies, I welcome opportunities for collaboration either as a co-author or as a consultant to help support the continued development of the EP Toolkit.

Release Notes:

1) Fixed crash when using Add button in Edit function to add a single cell.
2) Fixed crash in View Waves function when minimum voltage equals the maximum voltage.  
3) Fixed when clicking on Views channels, can only get expanded channel window if one clicks along the very top edge for FFT data.
4) Fixed upper and lower amplitude/power being changed immediately after new values manually entered into the View panel for FFT data.
5) Fixed min and max values displayed in Views pane divided by Hz bin width rather than sqrt(Hz bin width) when set to amplitudes.
6) Now allows power scaled data to be displayed as amplitudes.
7) Amplitude FFT data are represented in complex form.
8) Consolidated spectral unit controls so just four options (cm, am, pw, dB).
9) Fixed power-scaled data in View Waves function divided by square root of Hz bin width rather than Hz bin width.
10) Now allows power scaled FFT data to be output as amplitudes.
11) Fixed Matlab crash in View function when only FFT erpimages are being displayed.  
12) Fixed erpimages of FFT data displaying as blank.
13) If minimum value in View pane is zero, in dB scaling will scale to -2 instead of –inf to maintain useful range.
14) Eliminated option to transform data in power rather than amplitude form to avoid potential confusion.
15) Fixed crash when performing frequency transform on continuous data files.
16) Now doubles amplitude of spectral data when halving the bin size via the Edit function to correct for what it would have been and to ensure that subsequent spectral density scaling will be correct.
17) Fixed crash in View Waves when unable to generate unique labels for the four datasets.
18) Fixed no support for frequency PCA of continuous data in min and max values for View pane.
19) Added improved Scan pane for viewing and editing continuous data.
20) Disabled check to see if subtracting blink factors increased overall variance since semi-singularity correction seems to be controlling the noise problem.
21) Improved event marking of blink peaks.
22) Blink artifact channel now difference between upper and lower EOG channels.
23) Fixed couldn't read text files if in preference settings lastrow equals zero and firstrow is larger than 1.
24) Fixed jack-knife not calculated for first step of PCAs.
25) Fixed computation of peak pos and neg channels and peak channel polarity in Factors subpane of Edit function.
26) Fixed error when saving text files.
27) Fix to semi-singularity check for blinks, which was dropping too many channels.  
28) Dropped .power field.  Spectral data is now always maintained in amplitude scaling internally.

Joe


--------------------------------------------------------------------------------

Joseph Dien, PhD
Senior Research Scientist
Maryland Neuroimaging Center
University of Maryland, College Park
E-mail: jd...@ma... <mailto:jd...@ma...>
Cell Phone: 202-297-8117
http://joedien.com <http://joedien.com/>
------------------------------------------------------------------------------
_______________________________________________
Erppcatoolkit-support mailing list
Erp...@li...
https://lists.sourceforge.net/lists/listinfo/erppcatoolkit-support