We will conduct a series of masked priming experiments to investigate the morphological parsing of Tagalog infixed words. The existing masked priming literature has established that a brief and potentially unconscious (<60ms) presentation of a morphologically related prime (e.g., teacher) facilitates lexical decision to a target word (e.g., TEACH) as compared to when participants are presented with a completely unrelated prime (e.g., sugar) (see Rastle & Davis, 2008 for summary). Such findings have been taken as evidence supporting the existence of an early morphological parsing process in visual word recognition, in which the primes were rapidly parsed, and the constituent morphemes were subsequently activated, thereby facilitating the lexical decision when the stem is encountered again as a target. Masked priming effects have been observed in well-studied languages like English (Rastle & Davis, 2004, 2008; Rastle et al., 2000), Dutch (Drews & Zwitserlood, 1995), Spanish (Badecker & Allen, 2002), and French (Grainger et al. 1991). Outside of the Indo-European language family, masked morphological priming has been observed in less investigated languages like Setswana (Ciaccio et al., 2020), Basque (Duñabeitia et al., 2009), Japanese (Clahsen & Ikemoto, 2012; Fiorentino et al., 2016; Nakano et al., 2016), Korean (Kim et al., 2015), Hebrew (Boudelaa & Marslen-Wilson, 2005, 2011; Frost et al., 2000), and Turkish (Kirkici & Clahsen, 2013). These studies have established that morphological parsing could be a universal mechanism in language processing. There are several areas that need to be further explored: first, the vast majority of studies that demonstrated robust masked priming effects have focused on suffixation (e.g., Rastle & Davis, 2003; Rastle et al, 2004; Rastle & Davis, 2008; Clahsen & Ikemoto, 2012; Fiorentino et al., 2015), where the affixes can be easily located at the edge of the stem. How morphological parsing would take place when the affix is placed within the stem and when the form of the stem is disrupted remains under-investigated. This question can be explored using a language like Tagalog, where infixation is a dominant morphological process. Studies on Semitic languages like Hebrew and Arabic have addressed a similar question, asking whether masked priming effects will be found for a morphological process that interweaves a triconsonantal root and a phonological word pattern (e.g., the root zmr ‘anything to do with singing’ combined with the pattern _a_a_ forms zamar ‘a male singer’). Robust masked priming effects were found for prime-target pairs that share the same consonantal root (e.g., zmr-TIZMORET ‘anything to do with singing-ORCHESTRA’, Frost et al., 1997, 2000), suggesting that morphologically complex words in Hebrew are parsed into two abstract morphemes. However, such findings may not directly extend to Tagalog infixation. Hebrew’s templatic morphology maps the root and the pattern onto a consonant and vowel distinction, respectively, which can then be used as a cue to parse the morphologically complex words into consonantal roots and word patterns. This is not the case for Tagalog. It would be interesting to pin down what information Tagalog speakers use to track the stem and the infix in infixed words. As mentioned above, Tagalog involves disruption of morphemes whenever an infixed word is formed. How the morphological parser handles such disruption due to infixation has not been investigated. Although templatic morphology in Semitic languages also disrupts the linear phonological form of the root, it can be argued that there is no disruption to the critical prosodic representations of the morphemes, as existing theoretical and computational works have shown evidence that consonantal roots and vocalic patterns operate on separate prosodic tiers (McCarthy 1979, 1982; Dolatian & Rawski, 2020). Such separation of prosodic tiers was supported by the study of Gwilliams and Marantz (2015), where they found that Arabic speakers generate segmental phonological predictions in speech processing based on probabilities calculated over the dis-continuous consonantal root, rather than over the continuous whole word. On the other hand, Tagalog infixation is sensitive to prosody (see below), both the infix and stem involve both vowels and consonants, so both the linear and prosodic stem representations are disrupted by the infix. Some results from studies of English visual word processing may be relevant: Christiansen et al. (2005) find that letter transpositions within one element of a complex prime (e.g., sunhsine) resulted in the same magnitude of masked morphological parsing of a constituent target (e.g. shine) as a prime with no transposition, but that transposition across a morpheme boundary (e.g., suhnsine) resulted in significantly less facilitation of the prime. They found that letter transposition crossing a real morpheme boundary (e.g., boasetr) similarly resulted in no masked priming effect, but letter transposition between and stem and affix in a pseudo-complex word like ‘bluster’ (e.g., blusetr) did not affect masked priming magnitudes. While Christiansen et al (2005)’s finding of a contrast between within and across morpheme letter transposition was replicated in Basque (Duñabeitia, Perea and Carreiras, 2007), but subsequent research in English (Rueckl & Rihim 2011, Diependaele et al 2013, Taft, Lee & Beyersmann 2018) and Basque (Perea and Carreiras (2006a) find no effect of morpheme boundary for letter transposition, and report significant masked priming effects for both within and across boundary transposition. Masked priming effects that are robust to transposition have also been reported in Japanese Katakana (Perea & Perez, 2009). In Hebrew, Velan and Frost (2007) report that letter transposition generally results in inhibition in masked priming, which they argue is due to the fact that Hebrew orthography typically only writes consonants, and that transposing the consonants in a Hebrew root generally results in a different existing root (e.g., the roots SLX ‘to send’, XLS ‘to dominate’, XSL ‘to toughen’ and LXS ‘to whisper’ are all distinct roots). They argue that the extent to which early morpho-orthographic processing is highly position sensitive must therefore depend on the grammar and orthography of the language. McCormick et al (2008) find that the various small orthographic discrepancies that commonly arise in English suffixation, such as letter doubling (eg. slipper rather than sliper), letter deletion (eg. computer, not computeer) and substitution (eg. happiness, not happyness) do not disrupt morphological priming, despite somewhat obscuring the precise boundary between morphemes and breaking the perfect stem form identity between a complex word and its simplex stem. In contrast to the insensitivity to local, small disruptions to linear position found in these studies, Crepaldi et al. (2010; 2015) find that morphological parsing mechanisms are sensitive to larger differences in linear position. Specifically, they found no significant priming effects when the suffix -er appeared in the wrong position (e.g., sheeter vs. ersheet), while priming effects were maintained for stems regardless of the linear position (e.g., moonhoney vs. honeymoon), as stems appear in both edges of the word in English (e.g., dresser, overdress). In the case of Tagalog infixation, both the linear position of the stem and affix are altered (i.e., infixes are believed to be underlyingly prefixes, as will be further discussed below), and the finer grained linear sequence within the stem is also disrupted as seen in (1). ![enter image description here][1] Tagalog infixation offers an avenue to explore how much morphological parsing relies on preserving linear sequencing, alignment, and the form of the stem. Second, masked priming of prefixed words has been found to be less stable than for suffixed words. Some studies found masked prefix priming effects as robust as suffix priming effects (e.g., Marslen-Wilson et al., 1994; Giraudo & Voda, 2013; Beyersmann et al., 2016), or robust masked priming effects for prefixed words (e.g., Forster & Azuma, 2000; Chateau et al., 2002; Nikolova & Jarema, 2002; Diependaele et al., 2009; Kazanina, 2011; Ciaccio et al., 2020). On the other hand, there are also studies that found an asymmetry in the priming effects of prefixed and suffixed words, with suffixed words showing consistent priming effects regardless the lexicality and interpretability of the primes while priming of prefixed words was only observed when the primes are real words (e.g., in Korean, Kim et al., 2015). Interestingly, other priming studies have found the opposite direction of asymmetry, with prefixed word priming being obtained regardless of phonological opacity while suffixed word priming was only obtained if the prime and target are transparently phonologically related (e.g., in French, Marslen-Wilson et al., 2004). Several other studies have found this direction of more robust prefix priming effects than suffix priming effects (e.g., Grainger et al., 1991; Feldman & Laravee, 2001; Ciaccio et al., 2020). Investigating prefixation and suffixation in Tagalog in masked priming paradigms will speak to the debates of asymmetry in the morphological parsing of prefixed and suffixed words. Lastly, a recent MEG study by Wray et al. (2021) found that infixed words in Tagalog evoke a different pattern of early neural activity than either reduplicated prefixed Tagalog words, or prefixed or suffixed words in English, Greek or Japanese, as indexed by a response component associated with form-based morphological parsing. These findings could suggest an infix parsing mechanism that is distinct from prefix and suffix parsing mechanisms. There are two main reasons as to why this is a possibility: [1] from a theoretical linguistics perspective, scholars have argued that infixes are underlyingly prefixes that get relocated due to phonological readjustment rules (e.g., Kalin, 2015; Halle, 2001). For example, Halle (2001) on his work on Distributed Morphology (DM), used Tagalog infixation to show how infixes are underlyingly prefixes. Specifically, he argued that since, in the syntax proper, morphemes are nothing but complexes of syntactic and semantic features with no phonological exponents, it must be impossible within the syntax to infix /-in-/ before the first vowel of the Tagalog stem. He argued that this infix is underlyingly a CV prefix /ni-/. The prefix appears as an infix due to a rule of onset metathesis, which is a type of morphologically conditioned phonological rules. In Stockall and Gwilliams’ (under review) discussion of DM as a comprehension model, they propose that the parsing of morphologically complex words relies on unpacking such morphophonological readjustment rules, which will then add an additional step into the parsing of infixed words, while no such additional step will be needed for prefixed words that do not exhibit morphophonological readjustment rules (e.g., English, Greek, or Japanese prefixation and suffixation). [2] From a neurolinguistic perspective, studies have shown that the human brain is sensitive to this type of morphophonological rule during the early, form-based parsing stage (e.g., Fruchter et al., 2015; Wray et al., 2021). Particularly, the reliability of these morphophonological rules appear to be serving as cues during morphological parsing. What we do not know yet is whether the onset metathesis readjustment rules that trigger infixation are reliable enough for Tagalog speakers to use as cues to rapid, form-based morphological parsing, and whether using such cues requires additional processing resources, such as in the case of Tagalog reduplication and English irregular verbs allomorphy. Overall, our study aims to explore all these areas and inform our understanding of how current models of morphological parsing mechanisms in the visual domain can accommodate the language-specific properties of Tagalog infixed words. [1]: https://files.osf.io/v1/resources/5m3w9/providers/osfstorage/61e8185c53349e07565f9f88?mode=render