The output of bxh_eventstats (PREFIX_ is the output prefix specified when running bxh_eventstats):
PREFIX_LOG.txt – a log of all the external commands the program runs as well as some other data
PREFIX_LOG_EPOCHAVG.txt – a log from bxh_epochavg, which does most of the work
PREFIX_QUERIES.txt – a list of all the queries specified
PREFIX_baselineAvg.* – an average of all the points used as baselines for every trial
PREFIX_brainmask.* – if the “brainmask” option is specified, this is the mask
For each query with a label QUERYLABEL:
PREFIX_QUERYLABEL_avg.* – the raw averages of the time-locked epochs for each event that matched this query.
PREFIX_QUERYLABEL_avg_percent.* – same as above, but as a percentage of the mean image.
PREFIX_QUERYLABEL_n.* – the number of trials included (per voxel)
PREFIX_QUERYLABEL_std.* – stanard deviation of the time-locked epochs for each event that matched this query.
PREFIX_QUERYLABEL_std_percent.* – same as above, but as a percentage of the mean image.
PREFIX_QUERYLABEL_zerotmap.* – 4-D result of a t-test of the _avg image against a zero time course.
PREFIX_QUERYLABEL_cor.* – if a response template is specified using the “template” option, then this is the 3-D image representing the correlation between the time courses in the _avg image and the response template
PREFIX_QUERYLABEL_tmap.* – the t-statistic corresponding to the above _cor image.
For each comparison specified using the “tcompare QUERY1-QUERY2” option:
When running bxh_eventstats_combine, its output will include:
PREFIX_QUERYLABEL_avg.* – as above, though now these represent the mean of all the relevant _avg.* files.
PREFIX_QUERYLABEL_avg_percent.* – as above, though now these represent the mean of all the relevant _avg_percent.* files.
PREFIX_QUERYLABEL_tmap_zerottest.* and PREFIX_QUERY1_vs_QUERY2_tmap_zerottest.*: this represents a t-test of the relevant tmaps against zero.
PREFIX_QUERYLABEL_pmap_weightedz.* and PREFIX_QUERY1_vs_QUERY2_pmap_weightedz.*: this represents a weighted-z comparison of the relevant tmaps (see Whitlock, “Combining probability from independent tests: the weighted Z-method is superior to Fisher's approach”,
http://www.ncbi.nlm.nih.gov/pubmed/16135132 )