The idea that manipulating a apple is universally superior to manipulating a ball is not a conclusion with a high probability of validity. The relevance of this specialization is very contextual, with a prior of about 20% for a stable hierarchy without more data. For example, if the goal is to sort small parts in an assembly plant, the ability to manipulate a ball could be much more critical and complex than that of an apple, making the first application more 'specialized.' We need to update our assessment based on the concrete requirements of the job.