First, intelligent humans haven't been around for "millions of years."

The jury is still out on how intelligent our ancestors were, but the evidence suggests that they did not share our cognitive abilities. Tools remained fairly simple until only the last 200,000 - 300,000 years (at minimum), there's no evidence of abstract or symbolic thinking and expression (e.g., something like cave art) until Neanderthals, and literally brains were smaller and appear to have been less complex in some important ways (language centers, for example).

The earliest bipeds probably appeared around 6-8 million years ago, while the earliest evidence of tool use is only about 2.5 - 3 million years old at best (Olduvai Gorge and the Oldowan tool complex associated with Homo habilis and perhaps some Australopithecines contemporaneous with habilis). More complex tools appear beginning sometime around about 2 million years ago or a little less, and were probably made by Homo erectus and its cousins. Increased sophistication and refinement of the Acheulean complex (associated with erectus) continued through Homo heidelbergensis and we see increased diversity of tools with those and later Neanderthals.

Anatomically modern Homo sapiens doesn't seem to appear on the scene until about 300,000 years ago, and cognitive capability may not have tracked with anatomy. Until about 100,000 years ago, the human form may not have matched up with what we think of as the human mind.

As to the larger question of "independent civilizations," it more or less hinges on the development of agriculture and domesticated plants and animals. "Civilization" (which is generally viewed by most historians and anthropologists as "urbanized cultures") is characterized by large, sedentary populations. And such populations require stable, predictable, reproducible food resources. That's simply not possible until the development of domesticated plants and animals (and agriculture).

The earliest domesticated plants don't appear until about 11,000 - 12,000 years ago, give or take a few centuries, in SW Asia (Anatolia and the Fertile Crescent region). Fairly quickly after, other domesticated plant species (and animals) appear elsewhere around the world in various regions, largely consistent with where the earliest city states and empires appeared, both in the Old and the New Worlds.

Is it possible that prior civilizations had in fact existed but because they may have been tens or hundreds of thousands of years old little or no remnants, artifacts, etc. survive?

Many things are possible but the fact that there has been zero evidence of any such civilizations found, while there is plenty of evidence of peoples living tens and hundreds of thousands of years ago, suggests that such civilizations never existed.

It sounds like you're referring to the origin of civilizations within history (that is, within the historical or written record), rather than civilization. Civilizations have of course occurred throughout prehistory on every continent you can find people. Anatomically-correct humans only for the last couple hundred thousand years, with an origin for the development of language postulated anywhere from 40,000 to 100,000 years ago. You can't easily have a civilization without language, or at least I don't know what that would look like.

I can only speak to the development of writing, which occurred only four times independently - so far as we know. These accepted four innovations and their approximate dates of origin are:

• Sumerian cuneiform - writing from ca. 3000 BCE; proto-writing ca. 8000 BCE (tokens). Originally logographic, then logo-syllabic. Later adapted to Akkadian/Assyrian systems (logo-syllabic), Hittite (logo-syllabic), Elamite (syllabic), Old Persian (alphabetic), and a handful of other languages in Mesopotamia.

• Egyptian hieroglyphs - writing from ca. 2900 BCE; proto-writing ca. 3200 BCE (relationship to a few samples on pottery dating to several hundred years before this has been postulated, but are not conclusively proto-writing). Attempts to assign Proto-Sinaitic alphabet/abjad/brahmic syllabary scripts to hieroglyphs remain unsubstantiated. It is generally accepted, however, that these scripts ultimately derive from some sort of hieratic/cursive or shorthand form of either hieroglyphs or cuneiform.

• Chinese logograms - writing from ca. 1200 BCE; proto-writing from various sites ranging from ca. 2500 BCE. Several sites dating 1000 years earlier, and one dating 4000 years earlier, have been suggested but not proven to be proto-writing (Banpo, and Jiahu).

• Mesoamerican hieroglyphs - Olmec, Zapotec, and Mayan logo-syllabic systems with various contested order of origination. Earliest deciphered writing is from 300s BCE; earliest extant signs, not yet identified as proto-writing, from 900s BCE.

• Other systems - Several systems remain undeciphered, such as the Indus Valley script, Linear A, and Cretan hieroglyphs, but in each case in the absence of further information, linguists take these to be either derivative scripts of existing systems, or examples of proto-writing. Linear A, although likely unique from either Egyptian or Mesopotamian systems, far post-dates the origin of these scripts meaning trade existed between Crete and peoples with a long tradition of writing. This suggests that the idea of writing was transmitted to them, if not use of an existing script (a situation paralleled many times, such as with the creation of the Armenian script, or Luwian hieroglyphs, which therefore do not constitute unique writing any more than Klingon is). Rongorongo, a sophisticated script found on Easter Island, may constitute either writing or proto-writing, with no way to tell.

In the case of the Sumerian, Egyptian, and Chinese systems, it can be said actual writing originates within a few centuries into the Bronze Age for each respective region. For example Egypt's predynastic and early dynastic periods coincide with both writing and bronze metallurgy, and the same thing with the Shang dynasty in China.

Mesoamerican writing was developed in the Neolithic period for a region where metal working was rare. Bronze metallurgy existed in South American cultures in Argentina and also the Andes, but (other than quipu, a counting system in Peru and possible proto-writing) those regions did not develop writing.

There is good logic for bronze as a powerful condition for the development of a writing system, because of the increase in trade it makes available. Proto-writing has long been associated with trade (to show both denomination and symbol or seal of the transfer of ownership), and the first proto-writing in Egypt and Mesopotamia occurs on tokens and coinage. In Mesoamerica proto-writing appears calendrical in nature. Andean quipu appears to have been used exclusively for counting.