One way to check is to do a fold test on slickenline data. If the original fault surface is curved, the slip direction should not strongly depend on local fault plane orientation. If the fault surface is a folded plane, then the slikenlines should in principle have been parallel before folding. If direction of slip is parallel to the undulations of the fault surface, it is not possible to tell with this method, but a curviplanar fault is a simpler explanation.

In general, if a thrust surface is originally curviplanar, then the axis of this 'folded' curviplanar surface will typically be parallel to the thrust transport direction i.e. slickenline and thrust transport lineation data will be parallel to the Beta axis of the variable strike and dips of bedding surfaces on a stereoplot. Areas where thrust surfaces are raised are termed culminations, while areas where thrust surfaces are lower may be termed depressions. There may be systematic minor structural patterns around such structures over scales of ~100 km (see attached Xypolias and Alsop 2014 paper for further details).

Alternatively, if the deformation is later, it could have any orientation in relation to the thrust transport direction (i.e. it does not have to be parallel). In addition, later folding could develop anywhere, and would not necessarily be restricted to the vicinity of the thrust fault.

I agree.   The two suggestions above are different ways of answering the same question:  Is the slip direction on the fault (shown by slickenlines) parallel to the axis of folds in the fault surface (shown by the intersection of fault planes)?  If yes, the fault surface was originally curviplanar.

I also agree. I have mapped areas exhibiting  oblique slip thrusting in which fold axis making smaller  angles with transport direction. There are  situations where the fold  axis is almost normal to the transport direction. If the thrust slab is folded, the attitude of the stretching lineations are to be studied carefully for ascertaining the transport direction. If the dip of the thrust plane increases near the surface, the thrust plane has to be originally curviplanar. Subsequently the thrust plane may get folded during further transport.  

Hello,

One other way to determine whether the thrust plane has been later folded would be to extend the area of study to the footwall. 

If the footwall is also folded in a concordant way to the "folds" that affect the thrust, then probably the whole unit has suffered from late folding. Notice that it would be very difficult, if not impossible, that the hangingwall and thrust fault would've been folded, and not at least the uppermost part of the footwall.

If the footwall is undeformed (unfolded), well that has to do with the mechanics of thrust generation and emplacement: the thrust may be curvilinear and the hangingwall will adapt itself to this geometry through fault-bend folds related to frontal, oblique and lateral ramps (or other types or folding). However, while all those processes take place in the hangingwall, the footwall may remain perfectly still, lacking most of the deformation.

Hope this helps,

Cheers,

Manuel