Search

We report the results of two speeded acceptability judgment experiments in Turkish. We hypothesized an alternative explanation for agreement attraction effects in Turkish that is based on shallow processing. Our findings contradict our hypothesized form-driven processing strategy and support an account of agreement attraction based on the use of abstract linguistic features, rather than mere form.

Previous studies have shown that speakers may find sentences violating subject-verb agreement grammatical when the sentence contains a feature-matching noun phrase. This so-called agreement attraction effect has also been found in genitive possessive structures such as 'the teacher's brother' in Turkish (Lago et al., 2019), which is in contrast with its absence in similar constructions in English (Nicol et al., 2016). This discrepancy has been hypothesized to be a result of the association between genitive case marking and subjecthood in Turkish, but not in English. In the present research, we test an alternative explanation in which Turkish number agreement attraction effects are due to a potential confound in Lago et al.'s experiment, as a result of which subject head nouns were locally ambiguous between the possessive and the accusative case. We hypothesized that this ambiguity may have inhibited the availability of the head noun as an agreement controller as the accusative is a non-subject case in Turkish. To test this hypothesis, we conducted a speeded acceptability judgment experiment and our results suggest that case-ambiguity does not play a role in agreement attraction, and thus lends credibility to the claim that genitive noun phrases may function as attractors in Turkish due to the association between genitive case and subjecthood.

We examined the effects of argument-head distance in SVO and SOV languages (Spanish and German), while taking into account readers' working memory capacity and controlling for expectation (Levy, 2008) and other factors. We predicted only locality effects, that is, a slowdown produced by increased dependency distance (Gibson, 2000; Lewis and Vasishth, 2005). Furthermore, we expected stronger locality effects for readers with low working memory capacity. Contrary to our predictions, low-capacity readers showed faster reading with increased distance, while high-capacity readers showed locality effects. We suggest that while the locality effects are compatible with memory-based explanations, the speedup of low-capacity readers can be explained by an increased probability of retrieval failure. We present a computational model based on ACT-R built under the previous assumptions, which is able to give a qualitative account for the present data and can be tested in future research. Our results suggest that in some cases, interpreting longer RTs as indexing increased processing difficulty and shorter RTs as facilitation may be too simplistic: The same increase in processing difficulty may lead to slowdowns in high-capacity readers and speedups in low-capacity ones. Ignoring individual level capacity differences when investigating locality effects may lead to misleading conclusions.

A key problem in automatic annotation of historical corpora is inconsistent spelling. Because the spelling of some word forms can differ between texts, a language model trained on already annotated treebanks may fail to recognize known word forms due to differences in spelling. In the present work, we explore the feasibility of an unsupervised method for spelling-adjustment for the purpose of improved part of speech (POS) tagging. To this end, we present a method for spelling normalization based on weighted edit distances, which exploits within-text spelling variation. We then evaluate the improvement in taging accuracy resulting from between-texts spelling normalization in two tagging experiments on several Early New High German (ENHG) texts.

Psikodilbilim içindeki araştırma alanlarından biri insanların cümleleri nasıl anladığına odaklanır. Cümle anlama çalışmalarında gösterilmiştir ki dil kullanıcıları cümleleri bazen 'yüzeysel' bir şekilde işlemlemektedir (Barton & Sanford, 1993; Ferreira, 2003), standart dışı cümleler sık sık yanlış anlaşılmaktadır (Ferreira, 2003), çevrimiçi cümle anlamadaki yanlış anlamalar eksik yeniden-çözümlemeden ötürüdür (Christianson et al., 2001). Bu bulgular Yeterince-İyi (Ferreira, 2003; Ferreira, Bailey & Ferraro, 2002; Ferreira & Patson, 2007; Sanford & Sturt, 2002; Swets, Desmet, Clifton & Ferreira, 2008; Townsend & Bever, 2001) işlemleme yaklaşımına öncü olmuştur. Bu yaklaşımda, cümle anlama sisteminin `sezgisel` olarak adlandırılan stratejiler geliştirebildiği ve bu kısayolları (Swets et al., 2008) kullandığı varsayılmaktadır. Bu çalışmanın merkezinde adların canlı olma özellikleri üzerinden geliştirilen sezgiselin cümle işlemlemedeki rolü vardır. Bunu test etmek için, bölgesel olarak iki anlamlı olan yapılar içeren cümleler kullanılmaktadır. Kendi-hızında okuma deneyinin sonuçları göstermektedir ki insanlar aynı iki anlamlı yapıya, yapıyı oluşturan adların canlılık özelliğine göre farklı şekilde davranabilmektedir. Ayrıca, bölgesel olarak iki anlamlı olan yapının netleştirilmesi de ad dizisindeki adların canlılık özelliklerinden etkilenebilmektedir. One of the main research areas in psycholinguistic focuses on how people understand sentences. In sentence comprehension studies, it has been shown that language users sometimes engage in 'shallow' processing (Barton & Sanford, 1993; Ferreira, 2003), that non-canonical sentences are often misinterpreted (Ferreira, 2003), that misinterpretations during online comprehension are due to incomplete reanalysis (Christianson et al., 2001). These findings have given way to the Good-Enough processing approach (Ferreira, 2003; Ferreira, Bailey & Ferraro, 2002; Ferreira & Patson, 2007; Sanford & Sturt, 2002; Swets, Desmet, Clifton & Ferreira, 2008; Townsend & Bever, 2001). In this approach, it is assumed that the sentence comprehension system can develop parsing strategies that are called `heuristics` and make us of these `shortcuts` (Swets et al., 2008). At the center of the present study lies the role of heuristics based on animacy in processing sentences. To test this, Turkish sentences including a local ambiguity are used in this study. Results of a self-paced reading experiment show that people may treat the same ambiguous sequence differently based on the animacy of the noun phrases that make up the sequence. In addition, the disambiguation of the locally ambiguous sequence may also be affected by the animacy of the noun phrases in the sequence. 130

Iambik-Trokaik Yasa (Hayes, 1995), ritmik gruplamanın altında yatan evrensel bir mekanizma olarak öne sürülmüştür. Bu yasaya göre, ses düzeyi veya ses perdesi bakımından değişen ikili ses grubundan oluşan bir dizideki sesler, 'trokaik' olarak, yani kuvvetli olan önce gelecek şekilde gruplanır (Langus et al., 2016). Süre bakımından değişen ikili ses grubundan oluşan bir dizideki sesler ise 'iambik' olarak, yani kısa olan ses önce gelecek şekilde gruplanır (Hayes, 1995). Bu konuda yapılmış olan çalışmalarda, konuşma seslerinden (Crowhurst & Teodocio Olivares, 2014; Langus ve ark., 2016) ve sentetik seslerden (Iversen ve ark., 2008; Molnar ve ark., 2016) oluşan dizilerin gruplanmasında iambik-trokaik yasaya uymayan sonuçlar elde edilmiş, ve bu sonuçlar, anadilin gruplamaya etkisi olarak açıklanmıştır. Mevcut çalışmada, anadili Türkçe olan katılımcılara sentetik seslerden (Deney 1A ve Deney 1B) ve hecelerden (Deney 2) oluşan diziler dinletip gruplama tercihleri incelenmiştir. Gruplamada anadilin etkisi olmadığını gösteren sonuçlar, anadili Türkçe olan kişilerin, Türkçe'deki vurgunun tahmin edilebilir olması nedeniyle, vurgu değişimlerine karşı algısal bir hassasiyet geliştirmemiş olmaları ile açıklanmıştır. Anadilindeki vurgu tahmin edilebilir olan kişilerin, vurgu değişimlerine karşı düşük hassasiyetleri olduğunu raporlayan çalışmalar (Dupoux ve ark., 1997; Peperkamp & Dupoux, 2002; Rahmani ve ark., 2015; Lu ve ark., 2018) bu sonucu destekler niteliktedir. The Iambic-Trochaic Law (Hayes, 1995) was proposed as a universal mechanism governing rhythmic grouping. It predicts that sequences of sound events are grouped trochaically (strong-weak) when they alternate in intensity or pitch, and iambically (short-long) when they alternate in duration (Langus et al., 2016). Previous studies have found native language effects on grouping of both linguistic (Crowhurst & Teodocio Olivares, 2014; Langus et al., 2016) and non-linguistic (Iversen et al., 2008; Molnar et al., 2016) sounds, as they revealed grouping preferences incompatible with the predictions of the ITL. The present study investigated native language effects on rhythmic grouping by presenting native Turkish speakers sequences of tones (Experiment 1A and Experiment 1B) and syllables (Experiment 2). Lack of native language effects on grouping as revealed by the results, was explained by the stress 'deafness' (i. e., reduced sensitivity to stress deviations) exhibited by Turkish speakers, under the assumption that Turkish has predictable, fixed stress. That speakers of languages with fixed, lexically unmarked stress were reported to exhibit similar insensitivities to stress deviations (Dupoux et al., 1997; Peperkamp & Dupoux, 2002; Rahmani et al., 2015; Lu et al., 2018) supported this conclusion. 102