Mindless reading revisited: an analysis based on the SWIFT model of eye-movement control

A computational modeling approach to investigating mind wandering-related adjustments to gaze behavior during scene viewing

Research on gaze control has long shown that increased visual-cognitive processing demands in scene viewing are associated with longer fixation durations. More recently, though, longer durations have also been linked to mind wandering, a perceptually decoupled state of attention marked by decreased visual-cognitive processing. Toward better understanding the relationship between fixation durations and visual-cognitive processing, we ran simulations using an established random-walk model for saccade timing and programming and assessed which model parameters best predicted modulations in fixation durations associated with mind wandering compared to attentive viewing. Mind wandering-related fixation durations were best described as an increase in the variability of the fixation-generating process, leading to more variable-sometimes very long-durations. In contrast, past research showed that increased processing demands increased the mean duration of the fixation-generating process. The findings thus illustrate that mind wandering and processing demands modulate fixation durations through different mechanisms in scene viewing. This suggests that processing demands cannot be inferred from changes in fixation durations without understanding the underlying mechanism by which these changes were generated.

Do your eyes give you away? A validation study of eye-movement measures used as indicators for mindless reading

Identifying eye-movement measures as objective indicators of mind wandering seems to be a work in progress. We reviewed research comparing eye movements during self-categorized episodes of normal versus mindless reading and found little consensus regarding the specific measures that are sensitive to attentional decoupling during mind wandering. To address this issue of inconsistency, we conducted a new, high-powered eye-tracking experiment and considered all previously identified mind-wandering indicators. In our experiment, only three measures (reading time, fixation count, and first-fixation duration) positively predicted self-categorized mindless reading. Aside from these single measures, the word-frequency effect was found to be generally less pronounced during mindless-reading than during normal-reading episodes. To additionally test for convergent validity between the objective and subjective mind-wandering measures, we utilized eye-movement measures as well as thought reports, to examine the effect of metacognitive awareness on mind-wandering behavior. We expected that participants anticipating a difficult comprehension test would mind wander less during reading than would those anticipating an easy test. Although we were able to induce metacognitive expectancies about task difficulty, we found no evidence that these difficulty expectancies affected either subjectively reported or objectively measured mind wandering.

Linguistic processes do not beat visuo-motor constraints, but they modulate where the eyes move regardless of word boundaries: Evidence against top-down word-based eye-movement control during reading

Where readers move their eyes, while proceeding forward along lines of text, has long been assumed to be determined in a top-down word-based manner. According to this classical view, readers of alphabetic languages would invariably program their saccades towards the center of peripheral target words, as selected based on the (expected) needs of ongoing (word-identification) processing, and the variability in within-word landing positions would exclusively result from systematic and random errors. Here we put this predominant hypothesis to a strong test by estimating the respective influences of language-related variables (word frequency and word predictability) and lower-level visuo-motor factors (word length and saccadic launch-site distance to the beginning of words) on both word-skipping likelihood and within-word landing positions. Our eye-movement data were collected while forty participants read 316 pairs of sentences, that differed only by one word, the prime; this was either semantically related or unrelated to a following test word of variable frequency and length. We found that low-level visuo-motor variables largely predominated in determining which word would be fixated next, and where in a word the eye would land. In comparison, language-related variables only had tiny influences. Yet, linguistic variables affected both the likelihood of word skipping and within-word initial landing positions, all depending on the words’ length and how far on average the eye landed from the word boundaries, but pending the word could benefit from peripheral preview. These findings provide a strong case against the predominant word-based account of eye-movement guidance during reading, by showing that saccades are primarily driven by low-level visuo-motor processes, regardless of word boundaries, while being overall subject to subtle, one-off, language-based modulations. Our results also suggest that overall distributions of saccades’ landing positions, instead of truncated within-word landing-site distributions, should be used for a better understanding of eye-movement guidance during reading.

Disentangling the mechanisms underlying infant fixation durations in scene perception: A computational account

The goal of this article is to investigate the unexplored mechanisms underlying the development of saccadic control in infancy by determining the generalizability and potential limitations of extending the CRISP theoretical framework and computational model of fixation durations (FDs) in adult scene-viewing to infants. The CRISP model was used to investigate the underlying mechanisms modulating FDs in 6-month-olds by applying the model to empirical eye-movement data gathered from groups of infants and adults during free-viewing of naturalistic and semi-naturalistic videos. Participants also performed a gap-overlap task to measure their disengagement abilities. Results confirmed the CRISP model's applicability to infant data. Specifically, model simulations support the view that infant saccade programming is completed in two stages: an initial labile stage, followed by a non-labile stage. Moreover, results from the empirical data and simulation studies highlighted the influence of the material viewed on the FD distributions in infants and adults, as well as the impact that the developmental state of the oculomotor system can have on saccade programming and execution at 6months. The present work suggests that infant FDs reflect on-line perceptual and cognitive activity in a similar way to adults, but that the individual developmental state of the oculomotor system affects this relationship at 6months. Furthermore, computational modeling filled the gaps of psychophysical studies and allowed the effects of these two factors on FDs to be simulated in infant data providing greater insights into the development of oculomotor and attentional control than can be gained from behavioral results alone.

The Preferred Viewing Location in Top-to-Bottom Sentence Reading

The preferred viewing location (PVL) is a robust finding in research on reading that when fixating on a word during normal sentence reading, readers tend to land slightly to the left of the center of the word. This is in contrast to the optimal viewing location (OVL) in single word recognition, which falls at the center of the word. The current study outlines the history of the PVL in eye-tracking since Rayner's 1979 original study, documenting the origins of these conflicting theoretical explanations. In addition, a new study is reported examining whether the PVL can be attributed solely to oculomotor error or a processing advantage by using an experimental manipulation that separates tracking direction (left-to-right reading) and landing position (left-to-right within a word). Sentences were presented to participants from the top to the bottom of a computer screen with one word per line while eye movements were recorded. In this presentation format, readers continued to land to the left of center, suggesting that the PVL in normal reading is not solely due to oculomotor error.

No Evidence for a Saccadic Range Effect for Visually Guided and Memory-Guided Saccades in Simple Saccade-Targeting Tasks

Saccades to single targets in peripheral vision are typically characterized by an undershoot bias. Putting this bias to a test, Kapoula [1] used a paradigm in which observers were presented with two different sets of target eccentricities that partially overlapped each other. Her data were suggestive of a saccadic range effect (SRE): There was a tendency for saccades to overshoot close targets and undershoot far targets in a block, suggesting that there was a response bias towards the center of eccentricities in a given block. Our Experiment 1 was a close replication of the original study by Kapoula [1]. In addition, we tested whether the SRE is sensitive to top-down requirements associated with the task, and we also varied the target presentation duration. In Experiments 1 and 2, we expected to replicate the SRE for a visual discrimination task. The simple visual saccade-targeting task in Experiment 3, entailing minimal top-down influence, was expected to elicit a weaker SRE. Voluntary saccades to remembered target locations in Experiment 3 were expected to elicit the strongest SRE. Contrary to these predictions, we did not observe a SRE in any of the tasks. Our findings complement the results reported by Gillen et al. [2] who failed to find the effect in a saccade-targeting task with a very brief target presentation. Together, these results suggest that unlike arm movements, saccadic eye movements are not biased towards making saccades of a constant, optimal amplitude for the task.

Eye-Movement Control in RAN and Reading

The present study examined the visual scanning hypothesis, which suggests that fluent oculomotor control is an important component underlying the predictive relationship between Rapid Automatized Naming (RAN) tasks and reading ability. Our approach was to isolate components of saccadic planning, articulation, and lexical retrieval in three modified RAN tasks. We analyzed two samples of undergraduate readers (age 17-27), we evaluated the incremental contributions of these components and found that saccadic planning to non-linguistic stimuli alone explained roughly one-third of the variance that conventional RAN tasks explained in eye-movements registered during text reading for comprehension. We conclude that the well-established predictive role of RAN for reading performance is in part due to the individual ability to coordinate rapid sequential eye-movements to visual non-linguistic stimuli.

The neural substrates of natural reading: a comparison of normal and nonword text using eyetracking and fMRI

Most previous studies investigating the neural correlates of reading have presented text using serial visual presentation (SVP), which may not fully reflect the underlying processes of natural reading. In the present study, eye movements and BOLD data were collected while subjects either read normal paragraphs naturally or moved their eyes through "paragraphs" of pseudo-text (pronounceable pseudowords or consonant letter strings) in two pseudo-reading conditions. Eye movement data established that subjects were reading and scanning the stimuli normally. A conjunction fMRI analysis across natural- and pseudo-reading showed that a common eye-movement network including frontal eye fields (FEF), supplementary eye fields (SEF), and intraparietal sulci was activated, consistent with previous studies using simpler eye movement tasks. In addition, natural reading versus pseudo-reading showed different patterns of brain activation: normal reading produced activation in a well-established language network that included superior temporal gyrus/sulcus, middle temporal gyrus (MTG), angular gyrus (AG), inferior frontal gyrus, and middle frontal gyrus, whereas pseudo-reading produced activation in an attentional network that included anterior/posterior cingulate and parietal cortex. These results are consistent with results found in previous single-saccade eye movement tasks and SVP reading studies, suggesting that component processes of eye-movement control and language processing observed in past fMRI research generalize to natural reading. The results also suggest that combining eyetracking and fMRI is a suitable method for investigating the component processes of natural reading in fMRI research.

Morphology of Primary Visual Cortex Predicts Individual Differences in Fixation Duration during Text Reading

In skilled reading, fixations are brief periods of time in which the eyes settle on words. E-Z Reader, a computational model of dynamic reading, posits that fixation durations are under real-time control of lexical processing. Lexical processing, in turn, requires efficient visual encoding. Here we tested the hypothesis that individual differences in fixation durations are related to individual differences in the efficiency of early visual encoding. To test this hypothesis, we recorded participants' eye movements during reading. We then examined individual differences in fixation duration distributions as a function of individual differences in the morphology of primary visual cortex measured from MRI scans. The results showed that greater gray matter surface area and volume in visual cortex predicted shorter and less variable fixation durations in reading. These results suggest that individual differences in eye movements during skilled reading are related to initial visual encoding, consistent with models such as E-Z Reader that emphasize lexical control over fixation time.

Oculomotor and cognitive control of eye movements in reading: evidence from mindless reading

In the present study, we investigated the influence of cognitive factors on eye-movement behaviors in reading. Participants performed two tasks: a normal-reading task, as well as a mindless-reading task in which letters were replaced with unreadable block shapes. This mindless-reading task served as an oculomotor control condition, simulating the visual aspects of reading but removing higher-level linguistic processing. Fixation durations, word skipping, and some regressions were influenced by cognitive factors, whereas eye movements within words appeared to be less open to cognitive control. Implications for models of eye-movement control in reading are discussed.

Optimal viewing position effects in the processing of isolated Chinese words

Previous studies have found that words are identified most quickly when the eyes fixate near the word center (the Optimal Viewing Position, OVP) in alphabetic languages. Two experiments were performed to determine the presence of OVP effects during the processing of isolated Chinese words. Participants' eye movements were recorded while they performed a lexical decision task. The results suggest that Chinese readers exhibit OVP effects and that the OVP tends to be the first character for 2-character words. For 3- and 4-character words, the OVP effects appear as a U-shaped curve with a minimum towards the second character. As fixations deviate from the OVP, word processing times increase at a rate of 30-70 ms per character, and fixation duration is strongly influenced by the initial viewing position. Moreover, the present study did not observe an I-OVP effect for first fixation durations nor a fixation-duration trade-off in two-fixation cases in the case of isolated Chinese words processing.

Effects of load on the time course of attentional engagement, disengagement, and orienting in reading

We examined how the frequency of the fixated word influences the spatiotemporal distribution of covert attention during reading. Participants discriminated gaze-contingent probes that occurred with different spatial and temporal offsets from randomly chosen fixation points during reading. We found that attention was initially focused at fixation and that subsequent defocusing was slower when the fixated word was lower in frequency. Later in a fixation, attention oriented more towards the next saccadic target for high- than for low-frequency words. These results constitute the first report of the time course of the effect of load on attentional engagement and orienting in reading. They are discussed in the context of serial and parallel models of reading.

Using reinforcement learning to examine dynamic attention allocation during reading

A fundamental question in reading research concerns whether attention is allocated strictly serially, supporting lexical processing of one word at a time, or in parallel, supporting concurrent lexical processing of two or more words (Reichle, Liversedge, Pollatsek, & Rayner, 2009). The origins of this debate are reviewed. We then report three simulations to address this question using artificial reading agents (Liu & Reichle, 2010; Reichle & Laurent, 2006) that learn to dynamically allocate attention to 1-4 words to "read" as efficiently as possible. These simulation results indicate that the agents strongly preferred serial word processing, although they occasionally attended to more than one word concurrently. The reason for this preference is discussed, along with implications for the debate about how humans allocate attention during reading.

Your mind wanders weakly, your mind wanders deeply: objective measures reveal mindless reading at different levels

When the mind wanders, attention turns away from the external environment and cognitive processing is decoupled from perceptual information. Mind wandering is usually treated as a dichotomy (dichotomy-hypothesis), and is often measured using self-reports. Here, we propose the levels of inattention hypothesis, which postulates attentional decoupling to graded degrees at different hierarchical levels of cognitive processing. To measure graded levels of attentional decoupling during reading we introduce the sustained attention to stimulus task (SAST), which is based on psychophysics of error detection. Under experimental conditions likely to induce mind wandering, we found that subjects were less likely to notice errors that required high-level processing for their detection as opposed to errors that only required low-level processing. Eye tracking revealed that before errors were overlooked influences of high- and low-level linguistic variables on eye fixations were reduced in a graded fashion, indicating episodes of mindless reading at weak and deep levels. Individual fixation durations predicted overlooking of lexical errors 5s before they occurred. Our findings support the levels of inattention hypothesis and suggest that different levels of mindless reading can be measured behaviorally in the SAST. Using eye tracking to detect mind wandering online represents a promising approach for the development of new techniques to study mind wandering and to ameliorate its negative consequences.

Eye movements in reading versus nonreading tasks: Using E-Z Reader to understand the role of word/stimulus familiarity

In this article, we extend our previous work (Reichle, Pollatsek, & Rayner, 2012) using the principles of the E-Z Reader model to examine the factors that determine when and where the eyes move in both reading and non-reading tasks, and in particular the role that word/stimulus familiarity plays in determining when the eyes move from one word/stimulus to the next. In doing this, we first provide a brief overview of E-Z Reader, including its assumption that word familiarity is the "engine" driving eye movements during reading. We then review the theoretical considerations that motivated this assumption, as well as recent empirical evidence supporting its validity. We also report the results of three new simulations that were intended to demonstrate the utility of the familiarity check in three tasks: (1) reading; (2) searching for a target word in embedded in text; and (3) searching for the letter O in linear arrays of Landolt Cs. The results of these simulations suggest that the familiarity check always improves task efficiency by speeding its rate of performance. We provide several arguments as to why this conclusion is not likely to be true for the two non-reading tasks, and in the final section of the paper, we provide a fourth simulation to test the hypothesis that problems associated with the mis-identification of words may also curtail the too liberal use of word familiarity.

Systematic influence of gaze position on pupil size measurement: analysis and correction

Cognitive effort is reflected in pupil dilation, but the assessment of pupil size is potentially susceptible to changes in gaze position. This study exemplarily used sentence reading as a stand-in for paradigms that assess pupil size in tasks during which changes in gaze position are unavoidable. The influence of gaze position on pupil size was first investigated by an artificial eye model with a fixed pupil size. Despite its fixed pupil size, the systematic measurements of the artificial eye model revealed substantial gaze-position-dependent changes in the measured pupil size. We evaluated two functions and showed that they can accurately capture and correct the gaze-dependent measurement error of pupil size recorded during a sentence-reading and an effortless z-string-scanning task. Implications for previous studies are discussed, and recommendations for future studies are provided.

Models of the reading process

Reading is a complex skill involving the orchestration of a number of components. Researchers often talk about a 'model of reading' when talking about only one aspect of the reading process (e.g., models of word identification are often referred to as 'models of reading'). Here, we review prominent models that are designed to account for (1) word identification, (2) syntactic parsing, (3) discourse representations, and (4) how certain aspects of language processing (e.g., word identification), in conjunction with other constraints (e.g., limited visual acuity, saccadic error) guide readers' eyes. Unfortunately, it is the case that these various models addressing specific aspects of the reading process seldom make contact with models dealing with other aspects of reading. Thus, for example, the models of word identification seldom make contact with models of eye-movement control, and vice versa. Although this may be unfortunate in some ways, it is quite understandable in other ways because reading itself is a very complex process. We discuss prototypical models of aspects of the reading process in the order mentioned above. We do not review all possible models but rather focus on those we view as being representative and most highly recognized. WIREs Cogn Sci 2010 1 787-799 This article is categorized under: Linguistics > Computational Models of Language Psychology > Language.

The effect of word position on eye-movements in sentence and paragraph reading

The present study explores the role of the word position-in-text in sentence and paragraph reading. Three eye-movement data sets based on the reading of Dutch and German unrelated sentences reveal a sizeable, replicable increase in reading times over several words at the beginning and the end of sentences. The data from the paragraph-based English-language Dundee corpus replicate the pattern and also indicate that the increase in inspection times is driven by the visual boundaries of the text organized in lines, rather than by syntactic sentence boundaries. We argue that this effect is independent of several established lexical, contextual, and oculomotor predictors of eye-movement behaviour. We also provide evidence that the effect of word position-in-text has two independent components: a start-up effect, arguably caused by a strategic oculomotor programme of saccade planning over the line of text, and a wrap-up effect, originating in cognitive processes of comprehension and semantic integration.

Eye Movements During Mindless Reading

Mindless reading occurs when the eyes continue moving across the page even though the mind is thinking about something unrelated to the text. Despite how commonly it occurs, very little is known about mindless reading. The present experiment examined eye movements during mindless reading. Comparisons of fixation-duration measures collected during intervals of normal reading and intervals of mindless reading indicate that fixations during the latter were longer and less affected by lexical and linguistic variables than fixations during the former. Also, eye movements immediately preceding self-caught mind wandering were especially erratic. These results suggest that the cognitive processes that guide eye movements during normal reading are not engaged during mindless reading. We discuss the implications of these findings for theories of eye movement control in reading, for the distinction between experiential awareness and meta-awareness, and for reading comprehension.

Readers Use Bayesian Estimation for Eye Movement Control

During reading, saccadic landing positions within words show a pronounced peak close to the word center, with an additional systematic error that is modulated by the distance from the launch site and the length of the target word. Here we show that the systematic variation of fixation positions within words, the saccadic range error, can be derived from Bayesian decision theory. We present the first mathematical model for the saccadic range error; this model makes explicit assumptions regarding underlying visual and oculomotor processes. Analyzing a corpus of eye movement recordings, we obtained results that are consistent with the view that readers use Bayesian estimation for saccade planning. Furthermore, we show that alternative models fail to reproduce the experimental data.