The authors write "For the elbow joints, four simulations were run each for Mononykus and Bannykus; using CCFs [Cartilage Correction Factors] calculated from both the radius and ulna cartilage of Gallus, and for the radius and ulna values of Coturnix ... This was done to reflect the full spectrum of possible articular cartilage thicknesses...", as Coturnix values are more limited than Gallus (taken from Holliday et al. 2010). This makes sense as domestic chickens are generally not given time to ossify fully before adult size is reached.
Under Results, they first say "Both simulations (table 1; MECgu, MECgr) using Gallus-derived CCFs for the elbow of Mononykus resulted in very large volumes of viable ROM [Range Of Motion], which are uninformative for interpretation of in vivo forelimb function so are not discussed further. Additional simulations run using the Gallus-derived CCFs, with a higher overlap threshold of 80% (significantly more conservative than thresholds used in other ROM studies, which often consider 50% overlap still to be in articulation [6,40]), also yield very large volumes of viable ROM, indicating that Gallus-derived CCFs are inappropriate for use in Mononykus. Therefore, the Mononykus simulation using CCFs derived from measurements of the ulna of Coturnix (table 1; MECcu) is selected as most appropriate for further discussion and for comparison with the results for Bannykus. Both Bannykus elbow simulations using CCFs derived from Gallus (table 1; BECgu, BECgr) retrieve similar results; the ulna CCF simulation results (table 1; BECgu) are selected for further discussion and comparison, and the radius CCF simulation is not discussed further. Henceforth, discussion of ‘6DOF’ or ‘cartilage’ ROM simulations for the elbow joints refer to the Mononykus-Coturnix ulna simulation MECcu, and the Bannykus-Gallus ulna simulation BECgu (table 1)."
So this is not my area of expertise, but wouldn't the large volume of viable range of motion using Gallus for Mononykus actually mean that we can't use these simulations to tell us much about whatever limits Mononykus had, since we can't say it DIDN'T have Gallus-level thickness? Instead the authors seem to want SOME answer, so they arbitrarily go with Coturnix-level thickness that provides more limitations. Which wouldn't be THAT bad except their conclusion is-
"We find that Bannykus has the capacity for various digging styles and generalized forelimb function, while Mononykus has more restricted motion and may have employed a highly specialized digging style." (from the abstract)
"Large extents of elbow flexion and extension are feasible in Bannykus; these are greater than in Mononykus."
"In Mononykus, however, the highly modified morphology of the forelimb appears to have resulted in a reduction in ROM at the shoulder and elbow joints. Elevation and depression of the humerus are lessened, as are extension and flexion of the elbow, indicating that forelimb function may have been more specialized."
So surprise surprise- Mononykus has more limited elbow mobility when constrained to have thinner elbow cartilage. The fact four of six mobility measurements using Gallus-level thickness are LESS restricted for Mononykus' elbow than Bannykus' would really seem to argue against a more restricted elbow in the former if anything. This is especially true given long-axis rotation and elbow adduction are actually MORE restricted in Gallus-level Bannykus than even Cotunix-level Mononykus!
Now of course the authors measured one other joint too- the shoulder joint. "For the alvarezsauroid shoulder joint, avian CCFs were insufficient to fulfil the overlap requirements of the APSE algorithm, so the offsets were iteratively increased to find the closest viable offset", which is fine. I would say this supports Mononykus having at least a more limited range of motion in the shoulder than Bannykus, except we have sentences like-
"In fact, the degree of [shoulder] retraction retrieved is anatomically unfeasible, taking the humerus underneath the scapular blade..."
"Similarly to Mononykus, biologically unfeasible humeral retraction is reconstructed..."
"A large degree of possible elbow abduction is retrieved; this is biologically unfeasible..."
As far as I can tell, there's no table listing ranges of motion when biologically unfeasible positions are excluded, so the provided numbers seem less useful. Overall I'm skeptical this type of analysis can tell us much, especially based on Table 5 of Carpenter and Wilson (2008), who found the following in Gallus for all soft tissues intact versus no skin or muscle-
Shoulder flexion- 40 degrees vs. 90
Shoulder extension- 10 degrees vs. 70
Elbow flexion- 5 degrees vs. 90
Elbow extension- 60 degrees vs. 70
So the in vivo range of motions looks to be less than what bone and cartilage would suggest by themselves, sometimes but not always by a huge margin. And considering just how different alvarezsauroid shoulder and elbow joints are from chickens', I don't think it would be safe at all to extrapolate and say "reduce the calculated shoulder flexion by 55%" or something. So did Mononykus have a smaller range of forelimb motion than Bannykus? I don't think this study can tell us.
Reference- Carpenter and Wilson, 2008. A new species of Camptosaurus (Ornithopoda: Dinosauria) from the Morrison Formation (Upper Jurassic) of Dinosaur National Monument, Utah, and a biomechanical analysis of its forelimb. Annals of Carnegie Museum. 76, 227-263.
Mickey Mortimer