You must be refering to spangenhelm type helmets. They were of Central Asian/steppe origin, and Romans adopted them from either Sasanians, Sarmatians, Goths or from any combination of them.
Of course, all we can do is make hypotheses why this change took place. But bear in mind that during the Principate oval or hexagonal shields (smaller than a rectangular scutum) already existed and were used by auxiliary forces and cavalry (see for example Trajan's column). The scutum was strictly used by legionary heavy infantry, and thus it was associated to a very specific type of combat and tactics (close order combat using pila and gladii).
We have two contemporary sources for Caracalla's and the early III century CE at large: Cassius Dio and Herodian. Cassius Dio was not a gossipy historian, but he belonged to a very aristocratic and traditionalist milieu and this shaped his attitudes and mores towards government and public life in general. His Roman History, written in Greek (he was a Bythinian Greek from Nicomedia) has not survived complete, and many parts are only known to us through later compilations in a radically abbreviated form put together in medieval times (Constantine VII Porphyrogenitus, John Xyphilinus and John Zonaras) or by quotations and references in works by later authors. His model was mainly Titus Livius, and his Roman History was a monumental history of Rome ab urbe condita, following Livius' example. The most interesting parts are precisely the final books, reporting events of which he was a direct witness (reigns of Commodus to Severus Alexander). Unfortunately, the books of this latter part of his work are the ones that have been preserved the worst. We lack whole parts, and the rest is exclusively known through medieval Byzantine recenssions.
Cassius Dio's testimony is exceptional because he belonged to the highest social and political rank of the empire. He was a senator, and twice consul. He was also an amicus of two emperors, Septimius Severus and Severus Alexander. This means that he was an insider of the Roman governing elite. But this also means that he was as conservative and reactionary as one might be, and his outlook of society and events in general was taken through a very narrow senatorial lens (the same one used by Livius, Pliny or Tacitus). The latter books of his work were written at an advanced age, and although the information they provide is invaluable, they were the writings of an embittered old man settling accounts with emperors and people in general against whom he had an axe to grind. The account of the reigns of Commodus, Septimius Severus, Caracalla, Elagabalus and Macrinus were written during his retirement in his native Nicomedia after Septimius Severus' death. He had been an amicus of Severus, consul under him, and been member f the emperor's private council. Caracalla dispensed with his services and an embittered Cassius Dio retired to Nicomedia to write. His narrative of Caracalla's reign was thus written from outside of Rome and of the ruling circles of the moment, and by a man who hated Caracalla for personal reasons and despised his personal behaviour: for an aristocrat like Cassius Dio, Caracalla's behaviour mixing with common soldiers was a show of populism of the worst kind and a disgrace to his social rank and post.
Taking this into account, Cassius Dio's information about Caracalla's reign must be taken with caution and never at face value.
Herodian is an alternative and less biased source. The problem is that the material's quality is much lower than with Cassius' work. Herodian wrote also in Greek (he was probably a Syrian from Antiochia) but he only rose to the rank of a minor functionary in the imperial administration; thus his work is definitely that of an outsider, and one that was at times not very well informed. His writing also leaves much to be desired, he lacked the education, intelligence and incisiveness of Cassius Dio, and his work shows these defficiences. But despite all this, his relative lack of bias makes for an invaluable source to compare with Dio's account.
There are no more contemporary sources. The rest of sources date from the IV century, and use material now lost. The most complete account is that of the Historia Augusta (HA), a very problematic work. But it's still an important source; it ofefrs biographies of all the emperors from Hadrian to the advent of Diocletian. It was written in Latin and follows the style of Suetonius' Lives of the Twelve Caesars, full of gossip and scandal. The problem is that unlike Suetonius, the author or authors of the HA (Scriptores Historiae Augustae, SHA) in many cases directly forge tales and all kind of falsehoods, making the HA a minefield for historians. It was originally written in Latin probably during the late IV century and draws from older works written also in Latin. For the 12 emperors from Nerva to Elagabalus, the source for the SHA was a work that is now lost, the collection of imperial biographies written by Marius Maximus. And this person was a contemporary of Cassius Dio, and also a senator and twice consul, which gives weight to his accounts; the problem is that unlike Cassius Dio, who followed Livius' and Tacitus' steps in writing "serious" history, Marius Maximus followed Suetonius, and went after gossip, scandal and amusement.
There are no more sources, other than archaeology, epigraphy, Egyptian papyri or writings of contemporary writers that made indirect mention of the events in the empire at large at the time (Philostratus, Tertullian, Sextus Africanus, etc.). We don't have even the full text of the momentous Constitutio Antoniniana, only a fragmentary payrus from Egypt with its mutilated Greek translation.
Severus and Caracalla raised very substantially the legionaries' pay and gave them important social privileges. But at the same time, the length of service and the strict discipline of the legions remained untouched. If the problem of lack of recruits persisted, this meant that these elements that had remained untouched were precisely the ones that old stock Roman citizens had a problem with, and it was not merely a money issue. Also, the creation of 3 new legions by Septimius Severus (I, II and III Parthicae), added to the new legiosn already raised by Marcus Aurelius (I, II and III Italicae) and the effects of the Antonine Plague probably did not help at all.
The Constitutio Antoniniana made possible for previous non-citizens of non-servile origins (called peregrini in Roman law) to make an extraordinary social jump: from being a non-citizen to equestrian rank, a process that usually had taken generations and lots of luck to achieve. It's perfectly possible to imagine that on a short term (which probably was the only thing that Caracalla was interested in, as he needed recruits for his oncoming eastern war) this combination effectively boosted the numbers of recruits for the legions.
