General information

We here provide several sets of parcels that can be used for defining subject-specific fROIs. The parcels are in the *.img format.

Please cite Fedorenko, E., Hsieh, P.-J., Nieto-CastaƱon, A., Whitfield-Gabrieli, S. & Kanwisher, N. (2010) when using the parcels provided here.

Key parcels for the Sentences>Nonwords (S>N) contrast

[These are the parcels used and discussed in Fedorenko et al., 2010.]

These parcels are shown in Figure 1 below.

Figure 1: Key S>N parcels.

 

In addition to these parcels, we provide a set of the same parcels defined in a slightly more conservative way. In particular, before parcellating the probabilistic overlap map, we thresholded the map by including a voxel only if at least 5/25 subjects showed a significant (at the same .05 FRD threshold) S>N effect in that voxel. These parcels are of course smaller than the parcels above (see Figure 2 below, for a comparison). These may be especially useful in cases where no subject-specific functional data are available to define individual fROIs and instead these parcels are used as group-level ROIs.

When might you want to use the original (larger) parcels? Larger parcels allow you to capture more of each individual subject's activations. Furthermore, these parcels may be more appropriate when working with populations where there might be larger variability in the locations of functional activations compared to the variability in the healthy adult population.

Figure 2: Key S>N parcels: original vs. more conservative versions.

 

Additional parcels for the Sentences>Nonwords contrast

Unlike the key parcels that have a non-zero intersection with at least 80% of individual subjects in our dataset (of n=25), these parcels have a non-zero intersection with 60-79% of individual subjects, which is still relatively high. Some of these (e.g., the left and right temporal pole regions) have been implicated in different aspects of linguistic processing, so may be worth examining in future studies.

 

Parcels for the functionally narrower contrasts

Although we did not see strong evidence for functionally specialized sub-regions within our Sentences>Nonwords parcels (see Fedorenko et al., 2010, Appendix D, for discussion), we wanted to provide parcels that were obtained by applying the GSS method to functionally narrower contrasts (in particular, Words>Nonwords, and Jabberwocky>Nonwords). These parcels are smaller than the S>N parcels but show response profiles similar to those of the S-N parcels that encompass these narrower parcels. However, it is possible that in future studies, these regions will show profiles that will be different in an interesting way from those of the larger S>N parcels, so we want to provide researchers with an option of using these parcels in addition to the main set of S>N parcels. (NB: The 3-condition version of the localizer will have to be used in order to use the W>N parcels, and the jabberwocky condition will have to be included in order to use J>N parcels.)

 

Parcels for the "multiple-demand (MD)" regions

Because our language localizer by design was more difficult in the control (nonwords) condition than in the sentences condition, the contrast Nonwords>Sentences can be used to identify regions in the frontal and parietal cortices that respond more during more cognitively demanding tasks (e.g., Duncan, 2001, 2010). So, a short (10-15 min) scan gives you both high-level language regions, and the "multiple demand (MD)" cognitive-effort-sensitive regions.

Another way to define MD regions is by averaging activity from a diverse set of tasks. Seven tasks were used in:

Fedorenko, E., Duncan, J., & Kanwisher, N. (submitted). Extreme domain-generality in specific regions of frontal and parietal cortex.

The full unthresholded volume can be downloaded in Nifti format here.

The units are t-statistics from contrasts that isolate cognitive demand, averaged across the seven tasks. To create a symmetrical volume, data from left and right hemispheres were averaged, then projected back to both hemispheres. The figure below shows the main foci of the multiple demand system when this volume is thresholded at a value of t>1.5.

The next figure shows a parcellation into regions of interest (7 of 27 clusters are visible), after thresholding at t>1.5. If a cluster contained multiple peaks that also survived t>2.7, then the larger cluster was subdivided, assigning each voxel to the nearest subregion at the higher threshold.

A volume that assigns unique numerical labels to each of these regions of interest can be downloaded here.

 

Parcels for the ventral visual regions (Julian et al., 2012)

http://web.mit.edu/bcs/nklab/GSS.shtml

 

HOME