Morphemes constitute the smallest meaning-bearing units of language that are combined to create complex words (e.g., kindness consists of the stem kind and the suffix -ness). Despite considerable behavioral evidence that morphologically complex written words are processed and represented via their constituent morphemes, the neural underpinnings of morphological processing remain poorly understood (Leminen, Smolka, Duñabeitia & Pliatsikas, 2018).

Research shows that we access word internal structure during visual word identification. For example, we see embedded words (e.g., HAT in THAT) and we are sensitive to morphological structure (DEAL–ER, BASKET–BALL). The exact mechanisms that govern this process, however, are not well established yet—what determines access to embedded stem semantics? What is the role of affixes in this process? To address these questions, we tested the activation of meaning of embedded word stems in the presence or absence of morphological structure using two semantic categorization tasks in Italian.

Humans are known to detect probabilistic regularities in learning materials. In language, this ability is particularly well-evidenced for the auditory modality. In two experiments, we tested if learners rely on statistical regularities also when processing text. Italian speakers performed a lexical decision task on a large set of letter strings, distinguishing between nonword strings and strings that were words in a novel language. Participants successfully detected that novel words complied with the statistics of the Italian lexicon, and used this information to perform the task.

It has been suggested that the visual word identification system identifies recurrent letter clusters (n–grams) as a bridge between letters and words. Some evidence may suggest, rather indirectly, that n–gram coding is a more general visual mechanism, used beyond reading to extract statistical regularities in smaller visual units to build larger visual units. Here we test this hypothesis directly, and explore the boundaries of the phenomenon—what type of visual object, if any, fails n–gram based statistical learning?