Why Italian monolingual dictionaries usually take complex and/or archaic examples from books instead of creating simpler examples, like the monolingual dictionaries of other languages do (eg Cambridge English dict, RAE Spanish dict, Larousse French dict, Duden German dict) ? That makes them much harder to read and understand.
While, as DaG has said, this tendency has been decreasing (and probably absent from more modern dictionaries, although I still remember it quite alive in the Devoto-Oli, the dictionary I used in school), it is true that Italian dictionaries traditionally prefer to give examples from literary sources. While I cannot conclusively provide a reason for this, I think that examining the history of dictionaries might shed some light.
The first dictionary of Italian (and almost the very first monolingual dictionary in any language), is the Vocabolario degli Accademici della Crusca. This dictionary was compiled explicitly with the intention of using it to define what the Italian language was. In fact those were years when the questione della lingua was raging. The Accademia was strongly influenced by the ideas of Lionardo Salviati, one of the great exponents of the purismo . Essentially they argued that the language was perfected in the Florentine of the thirteenth and fourteenth century, and that anyone wanting to write "good" Italian should try to conform to that norm as much as possible. This was different from the "classicist" position of Pietro Bembo (who took literary quality, and not age, as the main criterion for inclusion of a text) and in strong opposition with the ideas of Vincenzo Colli and Gian Giorgio Tríssino, who argued for a norm more tied to the contemporary usage.
As a consequence in the first edition of the Vocabolario published in 1612 contained for every word at least a sentence of an author of the accepted period, to prove with certainty that the word respected the desired criteria. This turned out to be impossible, and a few posterior or not Tuscan authors were allowed on a case by case basis. However some great authors like Torquato Tasso were explicitly excluded, as insufficiently "Tuscan" in the Crusca's opinion.
The Vocabolario was a smashing success, as it was vastly superior in coverage and presentation to the word-lists of the time. However the strong Florentinism provoked numerous criticisms, and in the third edition (1692) several authors, among which Tasso, were reintroduced.
Long story short, however, even if the Florentinism was attenuated the Vocabolario's success managed to make mainstream the essentially purist criterion: a word was part of the Italian language if and only if it was used in the (approved subset of) Italian literature, and contemporary usage is no guide at all. I conjecture that the immense influence of the Vocabolario is responsible for the tendency you remarked, even in other dictionaries.
 This name is a bit of an anachronism, since it was attached to the position only in the nineteenth century. Still, it's what the books I read use.
This question is very interesting, but is partially based on questionable premises.
1) Not all, and perhaps a minority of dictionaries “usually take complex and/or archaic examples from books”. Both in the past (Fanfani and Rigutini's Vocabolario italiano della lingua parlata, for instance) and today there are dictionaries that put contemporary, spoken Italian first.
Among those, there are of course works specifically addressing Italian (mostly schoolchildren) and foreign learners. But, most importantly, several major dictionaries have a strong focus on modern Italian. I'll mostly mention late Tullio De Mauro's work. The most accessible is his online dictionary; see for instance the entry fatto, which illustrates the word with usual sentences and phrases from everyday language, and news and politics language. De Mauro was a great linguist especially interested in the different levels of use of Italian words and phrases, who also edited the multi-volume Grande dizionario italiano della lingua dell’uso, which, since its very title, gives a priority to Italian as it is actually used. See also De Mauro's “vocabolario di base della lingua italiana”, a list of most frequent Italian words, coded according to their use and register.
Also – as per Gio's comment – (recent editions of) Hoepli, Garzanti, Sabatini-Coletti or Devoto-Oli have a focus on contemporary Italian (new Devoto-Oli's very subtitle is Il vocabolario dell'italiano contemporaneo).
2) On the other hand, one of the uses of dictionaries is to be able to read and study Italian literature, from before Dante to today, so they have to record those “complex and/or archaic examples” too. Especially the larger ones, such as the Grande Dizionario della Lingua Italiana, Treccani and partly Zanichelli ones would be lacking if they didn't record rare and historical uses too.
3) Finally – but I'm sure someone else will cover better this aspect – there are historical reasons while Italian culture used to be in general somewhat more inclined towards written transmission of culture, tradition, purism and great past authors than towards present-day use, but my impression is that this is something we are overcoming now.