Effects of word predictability on eye movements during Arabic reading

Contextual predictability influences both the probability and duration of eye fixations on words when reading Latinate alphabetic scripts like English and German. However, it is unknown whether word predictability influences eye movements in reading similarly for Semitic languages like Arabic, which are alphabetic languages with very different visual and linguistic characteristics. Such knowledge is nevertheless important for establishing the generality of mechanisms of eye-movement control across different alphabetic writing systems. Accordingly, we investigated word predictability effects in Arabic in two eye-movement experiments. Both produced shorter fixation times for words with high compared to low predictability, consistent with previous findings. Predictability did not influence skipping probabilities for (four- to eight-letter) words of varying length and morphological complexity (Experiment 1). However, it did for short (three- to four-letter) words with simpler structures (Experiment 2). We suggest that word-skipping is reduced, and affected less by contextual predictability, in Arabic compared to Latinate alphabetic reading, because of specific orthographic and morphological characteristics of the Arabic script.

Robust effects of working memory demand during naturalistic language comprehension in language-selective cortex

A standard view of human language processing is that comprehenders build richly structured mental representations of natural language utterances, word by word, using computationally costly memory operations supported by domain-general working memory resources. However, three core claims of this view have been questioned, with some prior work arguing that (1) rich word-by-word structure building is not a core function of the language comprehension system, (2) apparent working memory costs are underlyingly driven by word predictability (surprisal), and/or (3) language comprehension relies primarily on domain-general rather than domain-specific working memory resources. In this work, we simultaneously evaluate all three of these claims using naturalistic comprehension in fMRI. In each participant, we functionally localize (a) a language-selective network and (b) a ‘multiple-demand’ network that supports working memory across domains, and we analyze the responses in these two networks of interest during naturalistic story listening with respect to a range of theory-driven predictors of working memory demand under rigorous surprisal controls. Results show robust surprisal-independent effects of word-by-word memory demand in the language network and no effect of working memory demand in the multiple demand network. Our findings thus support the view that language comprehension (1) entails word-by-word structure building using (2) computationally intensive memory operations that are not explained by surprisal. However, these results challenge (3) the domain-generality of the resources that support these operations, instead indicating that working memory operations for language comprehension are carried out by the same neural resources that store linguistic knowledge.Significance StatementThis study uses fMRI to investigate signatures of working memory (WM) demand during naturalistic story listening, using a broad range of theoretically motivated estimates of WM demand. Results support a strong effect of WM demand in language-selective brain regions but no effect of WM demand in “multiple demand” regions that have previously been associated with WM in non-linguistic domains. We further show evidence that WM effects in language regions are distinct from effects of word predictability. Our findings support a core role for WM in incremental language processing, using WM resources that are specialized for language.

More than words: word predictability, prosody, gesture and mouth movements in natural language comprehension

The ecology of human language is face-to-face interaction, comprising cues such as prosody, co-speech gestures and mouth movements. Yet, the multimodal context is usually stripped away in experiments as dominant paradigms focus on linguistic processing only. In two studies we presented video-clips of an actress producing naturalistic passages to participants while recording their electroencephalogram. We quantified multimodal cues (prosody, gestures, mouth movements) and measured their effect on a well-established electroencephalographic marker of processing load in comprehension (N400). We found that brain responses to words were affected by informativeness of co-occurring multimodal cues, indicating that comprehension relies on linguistic and non-linguistic cues. Moreover, they were affected by interactions between the multimodal cues, indicating that the impact of each cue dynamically changes based on the informativeness of other cues. Thus, results show that multimodal cues are integral to comprehension, hence, our theories must move beyond the limited focus on speech and linguistic processing.

Context-Based Facilitation of Semantic Access Follows Both Logarithmic and Linear Functions of Stimulus Probability

Stimuli are easier to process when the preceding context (e.g., a sentence, in the case of a word) makes them predictable. However, it remains unclear whether context-based facilitation arises due to predictive preactivation of a limited set of relatively probable upcoming stimuli (with facilitation then linearly related to probability) or, instead, arises because the system maintains and updates a probability distribution across all items, as posited by accounts (e.g., surprisal theory) assuming a logarithmic function between predictability and processing effort. To adjudicate between these accounts, we measured the N400 component, an index of semantic access, evoked by sentence-final words of varying probability, including unpredictable words, which are never generated in human production norms. Word predictability was measured using both cloze probabilities and a state-of-the-art machine learning language model (GPT-2). We reanalyzed five datasets (n=138) to first demonstrate and then replicate that context-based facilitation on the N400 is graded and dissociates even among words with cloze probabilities at or near 0, as a function of very small differences in model-estimated predictability. Furthermore, we established that the relationship between word predictability and context-based facilitation on the N400 is neither purely linear nor purely logarithmic but instead combines both functions. We argue that such a composite function reveals properties of the mapping between words and semantic features and how feature- and word- related information is activated during on-line processing. Overall, the results provide powerful evidence for the role of internal models in shaping how the brain apprehends incoming stimulus information.

The inhibitory effect of word neighborhood size when reading with central field loss is modulated by word predictability and reading proficiency

For normally sighted readers, word neighborhood size (i.e., the total number of words that can be formed from a single word by changing only one letter) has a facilitator effect on word recognition. When reading with central field loss (CFL) however, individual letters may not be correctly identified, leading to possible misidentifications and a reverse neighborhood size effect. Here we investigate this inhibitory effect of word neighborhood size on reading performance and whether it is modulated by word predictability and reading proficiency. Nineteen patients with binocular CFL from 32 to 89 years old (mean ± SD = 75 ± 15) read short sentences presented with the self-paced reading paradigm. Accuracy and reading time were measured for each target word read, along with its predictability, i.e., its probability of occurrence following the two preceding words in the sentence using a trigram analysis. Linear mixed effects models were then fit to estimate the individual contributions of word neighborhood size, predictability, frequency and length on accuracy and reading time, while taking patients’ reading proficiency into account. For the less proficient readers, who have given up daily reading as a consequence of their visual impairment, we found that the effect of neighborhood size was reversed compared to normally sighted readers and of higher amplitude than the effect of frequency. Furthermore, this inhibitory effect is of greater amplitude (up to 50% decrease in reading speed) when a word is not easily predictable because its chances to occur after the two preceding words in a specific sentence are rather low. Severely impaired patients with CFL often quit reading on a daily basis because this task becomes simply too exhausting. Based on our results, we envision lexical text simplification as a new alternative to promote effective rehabilitation in these patients. By increasing reading accessibility for those who struggle the most, text simplification might be used as an efficient rehabilitation tool and daily reading assistive technology, fostering overall reading ability and fluency through increased practice.

A further look at ageing and word predictability effects in Chinese reading: Evidence from one-character words

Older adults are thought to compensate for slower lexical processing by making greater use of contextual knowledge, relative to young adults, to predict words in sentences. Accordingly, compared to young adults, older adults should produce larger contextual predictability effects in reading times and skipping rates for words. Empirical support for this account is nevertheless scarce. Perhaps the clearest evidence to date comes from a recent Chinese study showing larger word predictability effects for older adults in reading times but not skipping rates for two-character words. However, one possibility is that the absence of a word-skipping effect in this experiment was due to the older readers skipping words infrequently because of difficulty processing two-character words parafoveally. We therefore took a further look at this issue, using one-character target words to boost word-skipping. Young (18–30 years) and older (65+ years) adults read sentences containing a target word that was either highly predictable or less predictable from the prior sentence context. Our results replicate the finding that older adults produce larger word predictability effects in reading times but not word-skipping, despite high skipping rates. We discuss these findings in relation to ageing effects on reading in different writing systems.

Reading development at the text level: an investigation of surprisal and embedding-based text similarity effects on eye-movements in Chinese early readers

This paper describes the use of semantic similarity measures based on distributed representations of words, sentences, and paragraphs (so-called“embeddings”) to assess the impact of supra-lexical factors on eye movement data from early readers of Chinese. In addition, we used a corpus-based measure of surprisal to assess the impact of local word predictability. Eye movement data from 56 Chinese students were collected (a) in the students’ 4th grade and (b) one year later while they were in 5th grade. Results indicated that surprisal and some text similarity measures have a significant impact on the moment-to-moment processing of words in reading. The paper presents an easy-to-use set of tools for linking the low-level aspects of fixation durations to a hierarchy of sentence-level and paragraph-level features that can be computed automatically. The study is the first attempt, as far as we are aware, to track the developmental trajectory of these influences in developing readers across a range of reading abilities. The similarity-based measures described here can be used (a) to provide a measure of reader sensitivity to sentence and paragraph cohesion and (b) to assess specific texts for their suitability for readers of different reading ability levels.